Improving the resilience of supply chains against disruptions like the Covid-19 pandemic PDF Free Download

Name: Improving the resilience of supply chains against disruptions like the Covid-19 pandemic PDF
Author: Tommy Kirby

1 / 113

0 views•113 pages

Improving the resilience of supply chains against disruptions like the Covid-19 pandemic PDF Free Download

Improving the resilience of supply chains against disruptions like the Covid-19 pandemic PDF free Download. Think more deeply and widely.

Master’s Degree in Management -

International Management

Final Thesis

Improving the resilience of supply chains

against disruptions like the Covid-19

pandemic

Supervisor

Ch. Prof. Marco Tolotti

Graduand

Francesca Putti

Matriculation Number: 879105

Academic Year

2020/2021

This page was intentionally left blank

Table of Contents 
Introduction .......................................................................................................... 1 
The supply chain ............................................................................................ 3 
1 History ..................................................................................................... 3 
2 Definition and scope of supply chains .................................................... 6 
3 Closed-loop supply chains ..................................................................... 10 
4 Strategic Inventory Positioning  ............................................................ 13 
The Covid-19 pandemic and supply chains ................................................. 21 
1 The outbreak .......................................................................................... 21 
2 Impacts on supply chains ...................................................................... 24 
2.1 Demand Management ..................................................................... 28 
2.2 Supply management ....................................................................... 30 
2.3 Production management................................................................. 31 
2.4 Transportation and logistics management ..................................... 32 
2.5 Relationship management .............................................................. 34 
2.6 Supply chain-wide impact (causing impacts in internal, upstream, 
and downstream operations) ....................................................................... 36 
2.7 Financial management .................................................................... 39 
2.8 Sustainability management ............................................................ 40 
Building resilience  ....................................................................................... 42 
1 Risk management .................................................................................. 44 

1.1 Assessment ...................................................................................... 45 
1.2 Prevention ....................................................................................... 49 
1.3 Mitigation ........................................................................................ 50 
1.4 Recovery .......................................................................................... 51 
2 Resilience strategies .............................................................................. 52 
2.1 Assessing the supplier network structure ...................................... 52 
2.2 Stress testing ................................................................................... 56 
2.3 Diversifying sourcing ...................................................................... 58 
2.4 Increasing stock .............................................................................. 59 
2.5 Reshoring and domestic production .............................................. 60 
2.6 Technological advancements .......................................................... 62 
2.7 Increasing circularity ...................................................................... 66 
2.8 Altering production ......................................................................... 67 
Recommending resilience strategies through decision support systems ... 69 
1 Decision trees ......................................................................................... 69 
2 The general model ................................................................................. 71 
3 The model considering synergies .......................................................... 82 
4 Additional considerations ..................................................................... 91 
Conclusions ......................................................................................................... 95 
References ........................................................................................................... 97 
Works Cited....................................................................................................... 101 

Introduction

Supply chains have seen an exponential increase of complexity during the

decades, mainly due to the growing degree of globalization and the subsequently

ever-growing number of actors involved. Despite the benefits that a larger

network can bring to businesses, it has also increased their exposure to risks,

resulting in significant losses, both of economical and strategical nature, and

sometimes even shutdowns.

The recent worldwide pandemic of Covid-19 has proven the magnitude of this

interconnectedness by leaving no aspect of our life entirely unaffected. What in

the eyes of optimists seemed to be just a small inconvenience concerning a

limited area of China, quickly became a historically momentous event due to the

extremely linked world it arose in, which provided a fertile ground for a chain of

events resulting in the involvement of countless economies and industries.

This work aims to investigate the impacts that possible disruptions can have on

supply chains and how companies can improve their ability to overcome them.

Two main subject matters contribute to this research question: an analysis of

the case of the Covid-19 pandemic and its consequences on businesses,

particularly their supply chains, and an account of how the resilience of

companies can be strengthened.

Specifically, Chapter 1 explains how supply chains have evolved over the years,

provides some definitions, discusses its main goals and attributes, and then

focuses on some relevant and noteworthy features that characterize them.

Chapter 2 offers an account of how the pandemic caused by the Coronavirus first

broke out and analyzes, providing relevant examples, the specific impacts it has

caused to supply chains, breaking them down by areas of concern.

Chapter 3, on the other hand, focuses on increasing resilience by first

introducing why assessing risks and increasing resilience is a fundamental

activity for companies. It then formalizes how the risk management process

develops and how managers could apply it to reap its benefits. The chapter, then,

discusses some of the strategies that have been proposed in the literature or

implemented by companies to increase preparedness for potential risks or

mitigate their negative effects.

Finally, Chapter 4 offers a methodological approach that can be used by

decision-makers to reach a more informed evaluation on the best strategies to

implement in order to reduce the risk exposure of the company. Particularly,

basing the methodology on the theory of decision trees, it investigates the effect

of some of the strategies discussed in Chapter 3, considering both their

individual and joint implementation. Additionally, it provides some insights and

reflections on how the technique can be further customized to fit the features

and needs of the company under consideration.

1 The supply chain

The current chapter will present a broad overview of the basic concepts

surrounding supply chains with definitions and descriptions to provide a good

understanding of the subject matter, necessary for the full comprehension of the

next chapters and the issues they discuss.

1.1 History

Since ancient times, the main catalyst of the progress and growth of civilizations

has been the exchange of information, goods, and services with neighboring

cities and communities. Ancient Rome is the perfect example of great

organizational efforts: thanks to its advanced road system, postal service, and

fleet, the roman empire was able to become one of the first global forces. Trade,

however, is only as effective as the underlying network of operations is. The

ability to store and move goods efficiently and adequately is vital to the success

of a broad network of exchanges. (Zijm, et al., 2019)

According to the website Logmore (2019), “the first example of production with

a truly global supply network was most likely rum. The supply chain in this case

started with slaves who were moved from Africa to the Caribbean to grow the

sugarcane, which came from India, and it ended in distilleries in the US.”

The history of supply chain management starts with logistics. In 1911, Fredrick

Taylor wrote “The Principles of Scientific Management”, focusing his research

on the improvement of loading processes inside factories, which at the time were

manually carried out. After that, studies on logistics problems intensified in the

1940s during World War II to improve military operations. Particularly, during

this time and the following decade, the focus was on increasing efficiency

through the use of mechanization of the most labor-intensive processes, like

material handling, along with the improvement of space utilization in

warehouses. The trend of standardization followed through the 1950s, too, with

the rise of intermodal containers and intermodal transportation, critical for the

process of globalization that supply chains were undergoing (Veridian, 2018).

“Containerization not only increased the quantity of available space for goods,

but also increased the speed of the freight movement while decreasing the cost.

The speed increase came from more effective warehousing processes as well as

transport terminal efficiency.” (Logmore, 2019).

Figure 1 - Intermodal transport chain (The Geography of Transport Systems, n.d.)

With the formation of a dedicated national council in 1963, this decade saw a

shift in modes of transportation towards trucks, which then required new

scrutiny on warehousing, material handling, and freight transportation, all

summarized together with the term “physical distribution”. With the advent of

computers, during the two following decades, innovation in logistics increased

exponentially, introducing concepts like truck routing, spreadsheet- and map-

based planning, algorithms for prediction of issues, and vastly improving the

optimization of processes and inventory. Additionally, this period marks the

start of increased awareness towards the importance and impact on the bottom

line of logistics and supply chain management, leading managers to recognize

the opportunities that focusing on optimization could bring for the profits of the

company, thus leading to higher investments in new technologies and trained

professionals that could help achieve higher efficiency (Veridian, 2018). The

1970s also marked the commercial spread of barcoding, a technology patented

two decades earlier. “Its adoption was spurred forward by a standard requiring

an identifying number from the US National Association of Food Chains and

subsequent research showing large increases in profit from point scale scanning.

Once the barcode was adapted to become an internationally used standard, it

could be used from for monitoring of the supply chain both globally and

internationally.” (Logmore, 2019).

Following the success of Material Requirements Planning (MRP) systems

during the 1970s and 1980s, the 1990s saw the emergence of Enterprise

Resource Planning (ERP) systems, aimed at integrating the many different

databases companies adopted. This structural update allowed greater accuracy

and availability of data, leading to easier, more advanced, and more effective

planning. Towards the new millennium, the term “supply chain” started gaining

mainstream recognition, even though one of the first mentions can be seen in an

article of the financial times dating back to the 1980s (Kolenko, 2014). This

spread can be attributed to the rise of globalized manufacturing, mainly lead by

the growth in the Chinese market. This brought increased complexity of

operations due to larger networks of actors involved, connecting multiple

countries, legislations, industries, and capabilities. (Veridian, 2018)

Most recently, the diffusion of big data has allowed analytics to evolve and to

encourage the adoption of monitoring practices, especially real-time, inside

supply chains. This additional step in the management of supply chains allows

companies to meet the efficiency needs of stakeholders with the use of

technology. Additionally, this newfound supervision allows processes to be more

thoroughly scrutinized by authorities and the public, increasing the pressure on

firms to maintain high standards of sustainability and social responsibility.

(Logmore, 2019)

1.2 Definition and scope of supply chains

A supply chain is the set of activities and processes that allows raw materials to

be transformed into finished products. The concept of supply chain “applies to

the internal relationships between processes as well as the external relationships

between operations” (Slack, et al., 2016, p. 399). It includes every step from

sourcing supplies, to building the various components that form the product, to

the assembly of said components, and finally to the delivery of finished goods to

the final point of sale. (Zijm, et al., 2019)

Supply chain management, therefore, is the handling and coordination of all

activities, relationships, and flows of a supply chain, so that they work

seamlessly towards the achievement of the economic goals of the company.

Lummus and Vokurka (1999, p. 2) also state that “a key point [in supply chain

management] is that the entire process must be viewed as one system. Any

inefficiencies incurred across the supply chain (suppliers, manufacturing plants,

warehouses, customers, etc.) must be assessed to determine the true capabilities

of the process.”.

Figure 2 - Total integration required within the supply chain (Lummus & Vokurka, 1999)

One very predominant aspect is the involvement of multiple companies and

industries working together in a synchronized way to manage all activities

necessary for the chain to function properly, this is called an end-to-end supply

chain. So, in order to manage the supply chain, it is important to coordinate

suppliers, customers, and any external providers of services across different

channels. Managers from one company usually take an interest in the

performance of other companies, working together to ensure the entire chain

works successfully for the benefit of all parties (Lummus & Vokurka, 1999).

Because of the many relationships, interactions, and points between the

processes inside supply chains and between actors across different supply

chains, one glitch on one side of the chain quickly transforms into increasingly

large inaccuracies by the time it reaches the other end. This is known as the

“bullwhip” effect. The distortions can travel both upstream and downstream the

chain. The main causes of this are: (Lee, et al., 1997)

- Demand signal processing: there is information asymmetry between

retailers and upstream suppliers, causing the suppliers to assume

demand based on the retailer’s behavior.

- Rationing game: during shortages, suppliers will ration the number of

units available to order from each retailer, but this will cause retailers to

order in excess in fear of running out of supplies.

- Order batching: companies prefer periodically placing larger orders fewer

times.

- Price variations: agents will replenish stock when prices are low and delay

placing orders for as long as possible when prices are high.

One technicality that is often mistaken is the distinction between supply chains

and supply networks. Slack et al. (2016) define supply networks as “all the

operations that are linked together to provide products and services to end

customers. In large supply networks there can be many hundreds of supply

chains of linked operations passing through a single operation”.

The phase of supply chain management that concerns warehousing all goods, be

that raw materials, components, or finished products, and transporting them

from one node in the chain to another is called logistics. It is of fundamental

importance to the integrity of the supply chain, being the factor that determines

if final customers are able to receive products in time or not. Logistics

management includes inventory management, transportation, and fleet

management. (Zijm, et al., 2019)

According to Barbara Gaudenzi & Antonio Borghesi (2006), in order to be

working effectively and bring added value, supply chains must be focused on two

main objectives:

- Value offered to final customers and assurance of their satisfaction.

Companies should concentrate their effort on clients when designing

supply chains, in order to maximize the value offered and its perception.

Therefore, it is crucial to understand which elements contribute to the

creation and the increase of perceived value.

- Reactivity. Supply chains must be designed in a way that allows both a

quick flow of processes that generates final products in a short amount of

time and fast adaptation to possible changes and disruptions.

Additionally, Slack et al. (2016) define 5 different performance objectives for

supply networks. Three of these – quality, speed, and flexibility – are analogous

to the objectives already mentioned. The novel two are:

- Dependability: being able to reduce uncertainty inside the chain to avoid

inefficiencies and complications.

- Cost: aiming at reducing transaction costs such as inventories,

transportation between activities, the cost of locating suppliers, and

making agreements.

1.3 Closed-loop supply chains

Given the importance of sustainability, and its recent increased awareness, it is

worth dwelling on one particular aspect of supply chains. Additionally, the

importance of focusing on this matter will be highlighted in section 3.2.7, where

increasing circularity is proposed as a strategy for increasing the resilience of

supply chains.

The scope of the supply chain has been recently broadened to also encompass

the handling of returns: products that are shipped back to a previous point in

the chain because of defects or because they remain unsold. The handling of

these flows gives rise to the so-called closed-loop supply chains, where often the

objective is to upcycle materials or parts of the products, leading to cost savings

and more environmentally friendly practices. (Zijm, et al., 2019)

Closed-loop supply chains are composed of two stages: the forward chain, where

the initial production and distribution is carried out, and the reverse chain,

where the products are retrieved and upcycled. This flow is also referred to as

reverse logistics.

Van der Laan (2019) outlines five different aspects to consider when analyzing

a closed-loop supply chain:

- How the product is made: the materials, and how these materials are

combined, determine how easy or difficult it is to separate and recycle

them. For example, products entirely made of one type of plastic can

easily be melted down and made into something new.

- The cause for return, which is linked to the condition in which the item is.

It could be sent back because of defects, because it was never sold or

because it reached the end of its intended period of use. Better conditions

doubtlessly imply more options to recover value out of the product. For

example, a smartphone returned because of a cracked screen can be put

back into the chain with a simple replacement of the frontal glass.

Differently, a car returned under a scrapping policy can probably only be

scavenged for raw materials to recycle.

- The driver for the recovery, which impacts how the network is structured

and what its aim is. The main drivers are legislation, corporate social

responsibility, and economics.

- The way in which the main processes of the chain (acquisition, recovery,

and remarketing) are organized, and which is the focus of the

organizational effort.

- The actors involved in this last phase of the supply chain. In particular,

the distinction is made depending on who is the agent performing the

recovery: if it is the original manufacturer, then the chain is a closed

closed-loop supply chain, while if a third party not originally involved is

taking on this responsibility it is an open closed-loop supply chain.

These factors determine the type of configuration the closing flow of the supply

chain will assume. However, it is specified that warranty returns and similar are

not covered in these scenarios.

- When legislation is the main driver, the activities are usually carried out

by third parties monitored by governmental institutions. One example is

the scrapping of vehicles.

- When recovering materials to recycle is profitable, companies will often

undertake only specialized recycling due to the high investments required

by the facilities needed.

- When it is possible to create a value-added network that is profitable, this

will be done by either a third party or by the brand owner itself. In the

latter case, the manufacturer has the advantage of deep knowledge of the

product design and of being integrated into the distribution

infrastructure to facilitate collection. The best example of this is ink

cartridge refurbishing.

- When the goods can be used again with little to no repairs or intervention,

a collection and distribution network will form.

- When manufacturers are obligated to take back products, because of

customer protection laws and warranty, these returns are highly valuable

since the products are often almost brand new and can be resold after a

light refurbishment.

To evaluate if a reverse logistics network is feasible and profitable, and decide

which strategy to follow to best take advantage of it, companies must consider

the different values that can be obtained with it. These are:

- Value derived from cheaper sourcing of materials and avoiding fees for

waste disposal and environmental impact.

- Value from positive image gained by communicating recovery initiatives.

- Value from higher customer satisfaction due to increased services and

products offered.

- Value from the information collected while inspecting the products

returned, which can help improve both the products and the recovery

process.

Figure 3 - Simplified graphical representation of a closed-loop supply chain (Fathollahi-Fard, et al., 2018)

1.4 Strategic Inventory Positioning

Another fundamental matter when discussing the structure and activities of

supply chains is strategic inventory positioning. Since section 3.2.4 will examine

the increase of stock and warehouse inventory as a strategy to prevent or

mitigate possible disruptions to the continuity of business, it would be worth

formalizing how strategic inventory positioning operates and how it can be

beneficial for companies in aiding the pursuit of resilience.

Once orders are placed, items such as raw materials, components, and finished

products have to be stored while waiting for the next step. This accumulation is

called inventory. Material inventories can represent an important source of

frozen capital, therefore lowering inventory can free up financial assets now

available for other activities. However, minimizing it excessively can lead to

issues of order fulfillment and halting of flows. This issue will be discussed

further in the next pages.

There are five main reasons to keep an inventory:

- To minimize the impact of unexpected interruptions in supply or demand

(buffer inventory). This point will be discussed further later when

examining the impacts of the Covid-19 pandemic.

- To allow companies to produce different items using the same machinery

(cycle inventory).

- To manage the intrinsic asynchrony of supply chains (de-coupling

inventory).

- To handle planned fluctuations in demand or supply (anticipation

inventory).

- To cope with delays in the transportation of materials and components

(pipeline inventory).

Since processes rarely work in a synchronized way, inventories are necessary

between many points along a supply chain due to its uneven flow. “If there is a

difference between the timing or the rate of supply and demand at any point in

a process or network, then accumulations will occur. […] If an operation or

process can match supply and demand rates, it will also succeed in reducing its

inventory levels.” (Slack, et al., 2016)

This perspective explains the need for inventories, however even if they are not

strictly required because the supply chain is especially efficient and

synchronized, placing an inventory can still be a strategically advantageous

choice because it achieves what is called decoupling (as listed in the third bullet

point above), defined as “Creating independence between supply and use of

material. Commonly denotes providing inventory between operations so that

fluctuations in the production rate of the supplying operation do not constrain

production or use rates of the next operation” (Ptak & Smith, 2016). The points

in the chain chosen for the decoupling inventories manage to decrease the

negative impacts of disruptions by breaking the bullwhip effect, previously

explained, therefore mitigating variability.

“The selection of these points is a strategic decision that impacts the

performance of the supply-demand network in many regards: service, working

capital, expedite-related expenses, cash flow, and ultimately return on

investment” (Ptak & Smith, 2016). Therefore, where to position inventories is

an important strategic decision. The authors consider six factors when

determining the best possible position for purchased, manufactured, and

finished goods:

- Customer time tolerance: how much customers are willing to wait to

receive the service or good they requested before looking for an

alternative source.

- Market potential lead time: a lead time that would allow the company to

achieve more sales or charge a higher price.

- Sales order visibility horizon: how much awareness there is regarding

future orders, and therefore demand.

- External variability: the potential swings in demand and supply. It is

usually classified as high, medium, or low.

- Inventory leverage and flexibility: key points that contribute the most to

reducing lead time.

- Critical operation protection: some areas deeply influence the quality or

performance of the entire chain, and therefore must be safeguarded.

These factors are analyzed on a case-to-case basis to determine which ones have

the most impact and therefore should be prioritized.

Figure 4 - Benefits of decoupling points in reducing variability (Ptak & Smith, 2016)

Different considerations are relevant for strategic positioning regarding

distribution, where a balance must be sought between meeting market demands

quickly and maintaining access to the financial investments that an inventory

would require.

In a distribution network, the most significant source of instability is demand

variability, therefore the best solution would be to place warehouses at a hub

near the sourcing facility. Here volatility is lower because of the smoothing effect

that happens on variability when multiple events are aggregated. The amount of

smoothing can be computed mathematically through the coefficient of variance

formula.

Ptak & Smith argue that “the best place in a distribution network to mitigate and

manage demand variability is at a point of aggregation where there is less

inherent relative volatility. Yet this mathematical fact seems to be lost on the

people and organizations running the vast majority of distribution networks.

Many distribution networks are designed and managed in a way that prohibits

them from taking advantage of this concept.”

Many reasons to explain this apparent lack of rationality are presented:

- Shipping larger quantities of materials when organizing a shipment to

facilities down the chain increases the efficiency of transportation.

However, this leads to an oversupply of points down the chain while

leaving the main hub understocked.

- The performance of sourcing facilities is often measured on unit costs.

Bigger batches improve these metrics.

- The assumption that placing inventory closer to the point of consumption

offers the most benefits is often present within organizations.

- Sometimes this downstream distribution of stock is not due to the input

of the sourcing warehouse but rather of regional facilities overordering in

an effort to avoid a perceived scarcity.

- If the sourcing unit does not have a high storing capacity, the stock must

necessarily be relocated to downstream facilities.

These arguments often lead companies towards two situations. The first one is

when stock is not held at the main hub, which leads to cross shipments between

local facilities because there is a divergence between amounts needed and

amounts detained. It could also cause missed sales and lessen the benefits

mentioned above.

Alternatively, the main facility could be able to carry enough stock, but it is

simply not located strategically. As suggested before, the best position according

to Ptak and Smith (2016) is as close to the main sourcing unit as possible to

maximize availability while keeping lead times short and to avoid cross-

shipments between entities. Additionally, this strategy minimizes the bullwhip

effect, allows more efficient consumption of resources, and simplifies

production scheduling.

Figure 5 - Decoupled distribution network (Ptak & Smith, 2016)

However, this strategy is not always feasible for every company because of

constraints related to space, network structure, or others. Often the distribution

centers are several, making the above model inapplicable. In this case, one

solution could be converting one of the regional facilities to a hub, creating the

so-called “hub-and-spoke” model. The facility chosen for conversion should be

the one with the highest volume of business, the closest to necessities like

suppliers and means of transportation, or the one with the largest space

available. In the case of an international company, multiple regional hubs can

be designated, each covering one geographical area.

Additional advantages that come with this strategy are further decreasing

external variability, better space and freight utilization, and improved capacity

to meet large orders.

One final hybrid configuration is presented. Ptak and Smith (2016) argue that

this model is appropriate in the case of space scarcity at a sourcing unit since it

avoids the need for a full hub at that point. Additionally, “it focuses on

decoupling the variability between the sourcing unit and the distribution

network associated with slow-moving items [...] since their minimum quantity

requirements in relation to their usage rates often create significant imbalance

in the network and scheduling difficulties for the plant.” (Ptak & Smith, 2016, p.

101)

2 The Covid-19 pandemic and supply chains

The current chapter will provide an account of how the Covid-19 pandemic

started and its impact on the economy and specifically on supply chains, in order

to provide an understanding of the consequences that such a disruption implies.

2.1 The outbreak

On December 31st, 2019, news from the World Health Organization started

spreading regarding an unusually high number of pneumonia cases in Wuhan,

eastern China, caused by an unknown virus. Appearing to have started from

animals in the seafood market of the city, promptly shut down, the disease

quickly spread, with symptoms consisting mainly of fever, a dry cough,

shortness of breath, and fatigue, but presenting more severe manifestations

requiring hospitalization in weaker patients (Reynolds & Weiss, 2020).

On January 9th the first victim of the virus died of respiratory failure. Despite

the quarantine put over the areas initially involved, the virus reached across

borders and overseas in a matter of weeks, leading many countries to close

borders, put areas on lockdown and initiate emergency measures (CNN

Editorial Research, 2021). On March 11th, the WHO declared Covid-19 a

pandemic, which, as of August 2021, counts more than 200 million cases

worldwide and more than 4 million deaths (Worldometer, 2021).

Focusing on the economic impacts of the pandemic, experts at Washington

University have initially estimated an impact of over $300 billion on the global

supply chain, with effects that could last for over two years (Miller, 2020).

Financial markets were not safe from the negative effects of the pandemic,

seeing the sharpest falls since the economic crisis of 2008 (Carrick, 2020).

One of the hardest-hit industries was tourism due to local restrictions, closures,

and especially travel bans. “The decades-long travel boom has made us more

globally connected than ever before. But with no end in sight, the coronavirus

has made the industry both a vector for – and unfortunate poster child of – this

historic event.” (Turner, 2020)

Another sector heavily impacted globally was retail. With reduced opening

hours, lockdowns preventing people from physically going to stores, and

government decrees, retailers saw a great loss of income, which they were only

partly able to offset thanks to an increased focus on online operations. For

example, during January, February, and March 2020, European and UK retail

stores saw a decrease in customer visits of respectively 10.63%, 8.89%, and

41.43% when compared to the same months of 2019 (Santos, 2020).

Additionally, the pandemic had a heavy impact on global poverty, with

assessments claiming an increment of 97 million people in poverty caused by

the pandemic during 2020, a figure that is lower than initial estimates but still

remains a historically unprecedented increase. Even if poverty rates resume the

declining trend observed before the pandemic, millions of people will live in

poverty for many years to come due to the initial impact of Covid-19. (Mahler,

et al., 2021)

Figure 6 - Extreme poverty in 2015-2021, measured as the number of people living on less than $1.90 per day

(Mahler, et al., 2021)

The sanitary emergency caused by the Coronavirus can be classified as a crisis

for many reasons, according to the definition offered in their research by

Chowdhury and Quaddus (2016, pp. 710-711):

- “Organizational crisis is low probable but creates high impact.” Even

though the likelihood of a pandemic happening has increased in the last

decades (Yeung, 2019), this event can still be considered improbable. Its

effects, however, are of large scale: as of November 2020, the total cost of

the Covid-19 pandemic has been estimated to be between $8-$16 trillion

(Financial Express, 2020).

- “It threatens viability of the organization.” As will be discussed more

extensively in the next section, the Coronavirus severely impacted many

companies, sometimes even putting their survival at risk.

- “Stakeholders perceive crisis as personally and socially threatening.” The

change in consumption patterns right after the pandemic first hit proves

the public feared a high influence on their everyday life (Ernst and Young,

2020).

- “Cause, effect and means of resolution of crisis are ambiguous and may

shatter individual’s beliefs, values, and basic assumptions.” This was

demonstrated by the inability of many companies to effectively respond

to the emergency, either because of lack of planning or inadequacy of the

plans already formed (ISM, 2020).

- “Decision-making during crisis is constrained by time and cognitive

limitations.” The rapid pace of the pandemic along with its

unpredictability caused uncertainty about outcomes that ensued after

actions were taken, influencing the decision-making process (Gunessee &

Subramanian, 2020).

2.2 Impacts on supply chains

There are four main reasons, presented by Wilson (2020), why Covid-19 had a

bigger impact on supply chain compared to other disruptions like natural

disasters, restrictions, political unrest, and fluctuations in the economy:

- Geographic scope. Differently from extreme weather phenomena like

hurricanes, this pandemic had, and continues to have to this day, a global

reach, affecting countries and companies all around the world. This also

implies reluctance in sharing aid, supplies, and manpower, as usually

happens with regional events, because those elements will probably be

already needed locally.

- Industrial scope. Very often, in the case of disruptions in supply chains,

only a specific industry, or a few, is involved. For example, Hurricane

Harvey in 2017 had high negative impacts on refineries located on the

Gulf Coast. However, the ones in the rest of the USA were able to make

up the shortfall since they were not affected. The current pandemic has

affected almost every industry and company. Additionally, to the scarcity

of raw materials, many other essential items like sanitary supplies and

even food have seen shortages. Service providers, factories, and many

other activities that imply close human contact like people transport have

been forced to halt their business.

- Demand shame. Since manufacturers did not anticipate Covid-19

happening, they continued production as usual, particularly of costly

goods, reassured by the belief that high-end buyers would hardly alter

their consumption patterns even in the event of a slow year. However, as

a result of the pandemic, many companies have found themselves with

large volumes of high-value stock which they are unable to sell, leading to

frozen assets. Indeed, many responsible buyers are currently avoiding

expensive investments like cars. Additionally, for some luxury brands

halting business during such events represents a matter of public image:

after the tsunami of 2011 in Japan, Luis Vuitton closed all the stores in

the country, stating that “It just did not look right to be open and selling

luxury handbags when thousands of Japanese had just lost their homes”

(Wilson, 2020).

- Duration. Logically, most of the disruptions mentioned earlier, like

weather events, have a short-term scope and their consequences can be

overcome or at least partly dealt with within a few months. However,

pandemics last much longer because they naturally present relapses and

spikes. For instance, the Spanish Flu took almost two years to fade.

In their research, P. Chowdhury et al. (2021) offer a summary of the impacts of

the COVID-19 pandemic on supply chains classified by area, which will be used

here as a framework to later analyze each impact further, integrating relevant

literature.

Table 1 - List of impacts of the COVID-19 pandemic on supply chains (Chowdhury, et al., 2021)

Impacted area

Specific impact

Demand management

Demand spikes for essential products

Shortage of essential products

Loss of security with respect to essential items

Failure of on-time delivery

Declining demand for non-essential products

Ambiguity or difficulty in forecasting

Supply management

Shortage of material supply/supply-side

shock/supply disruption

Production

management

Production disruption and backlog

Reduced production capacity

Unavailability of workforce

Obsolescence and impairment of machinery

and capital assets

Transportation and

logistics management

Delays in transportation and distribution

Lack of international transportation/trade

Loss/lack of physical distribution channels

Shift of distribution and logistics pattern (offline

to online or blended)

Relationship

management

Reduced social interaction

Information ambiguity

Lack of supplier engagement/opportunistic

behavior

Supply chain-wide

impact (causing impacts

in internal, upstream,

and downstream

operations)

Ripple effect on all the operations involved in

supply chains

Supply chain collapse

Closure of facilities, including both companies’

production facilities and the facilities of supply

chain partners such as suppliers and

distributors

Financial management

Reduced supply chain financial performance

(e.g., loss/reduction of financial stability)

Reduced cash inflow

Sustainability

management

Lack of focus on social and environmental

sustainability practices/disruption of

sustainability initiatives

Threats to the health and safety of the

workforce

Contraction of the development of green and

low-carbon energy sources

Increase in waste

Increase in recyclable materials

2.2.1 Demand Management

- Demand spikes for essential products

- Shortage of essential products

- Loss of security with respect to essential items

- Failure of on-time delivery

- Declining demand for non-essential products

- Ambiguity or difficulty in forecasting

Following the Covid-19 outbreak, demand for many products greatly

transformed. Essential items like food and medical supplies quickly became

scarce because of panic buying, fueled by mass hysterics and the media’s

sensationalism, leading manufacturers to intensify production (Bagshaw &

Powell, 2020). On the other hand, nonessential items like leisure products or

goods requiring considerable investments, like cars or large appliances, saw a

drastic drop in sales, due to shifted priorities and lower disposable income

caused by a sudden rise in unemployment. (Zhu, et al., 2020)

According to an Ernst and Young survey (2020), more than two-fifths of

participants believed that their shopping habits would change drastically due to

the pandemic, with almost a third admitting to spending less overall, mainly as

a result of lower employment rates. The study divided people into four

categories, visible in the figure below: “cut deep” which represents people hit

hardest by the pandemic who are spending less, “stay calm, carry on” factoring

those who are continuing to spend as normal, “save and stockpile” which

includes pessimists worried about their families, and “hibernate and spend” who

are well-positioned to deal with negative effects and who are increasing

expenses. However, even amidst this divergence of behavior, the study found

that all categories decreased consumption of non-essential items, as illustrated

in the graphic below. These demands alterations undoubtedly caused

complications across the management of supply chains.

Figure 7 - Covid-19 impact on spending habits by segment (Ernst and Young, 2020)

2.2.2 Supply management

- Shortage of material supply/supply-side shock/supply disruption

Production and shipment of raw materials and components to be assembled

halted worldwide since borders and factories closed to contain the infection,

disrupting the production chain even in regions not yet affected or affected less

severely. (Zhu, et al., 2020)

Indeed, “in South Korea, automakers have shut down plants due to a lack of

component parts from China, implying that there are no alternative suppliers

ready to fill the gap” (Lierow, et al., 2020).

A whitepaper by GS1 US (2021) points out the two main implications that the

shortage of raw materials has entailed:

- High dependency on a few suppliers concentrated in limited geographical

areas.

- Asymmetry between the production that happens off-shore and the one

in factories closer to the final point-of-sale, implying increased difficulties

in predicting the availability of products to be sold, longer lead times, and

challenges during planning.

Impacts related to material shortages and unavailability of parts inevitably

compromise the integrity of the entire supply chain because of the

aforementioned ripple effect, therefore they will be better discussed in section

2.2.6.

2.2.3 Production management

- Production disruption and backlog

- Reduced production capacity

- Unavailability of workforce

- Obsolescence and impairment of machinery and capital assets

As found by a study by the Institute of Supply Management, Chinese

manufacturing facilities were operating with 56% of staff, causing significant

reductions in production capacity (ISM, 2020).

The closing of borders has decreased the number of migrant workers and

hindered the flow of commuters. One of the industries most affected by this is

agriculture, which heavily relies on seasonal job force. This matter raises worries

regarding possible food scarcity and the impacts of the entire food production

system, especially after the initial plans for the growth of locally sourced labor

did not deliver: “Despite much political noise around filling vacancies with

workers who lost their jobs in other sectors, only 150 workers started harvesting

jobs in UK agriculture as part of a scheme for which initially 50,000 UK workers

had signed up” (Trautrims, et al., 2020, p. 4). This unavailability of workforce

put stress on companies who experienced surges in demand or that have to hire

new workforce to fill in positions that are vacant due to layoffs needed at the

beginning of the pandemic, preventing them from running thorough selection

processes to assess skills and adequate qualifications because of the pressure to

keep up with the fast-paced nature of changes caused by the pandemic.

Additionally, the sudden and unplanned decrease in production resulted in the

obsolescence of materials and machinery before their usefulness was fully

exhausted, leading to a loss of utility and efficiency (Dente & Hashimoto, 2020).

2.2.4 Transportation and logistics management

- Delays in transportation and distribution

- Lack of international transportation/trade

- Loss/lack of physical distribution channels

- Shift of distribution and logistics pattern (offline to online or blended)

Since travel restrictions took effect in almost every country, fewer commercial

flights and trains were circulating. In addition to a decline in tourism, which

undoubtedly caused direct economic losses, this factor also implies reduced

opportunities to transport cargo, which previously could be loaded alongside

luggage in the aircraft hold. Moreover, transport trucks were limited by closed

borders on land and increased safety measures, leading to delays along the

entire process of transport, especially during customs clearance due to reduced

personnel and increased regards towards potential sources of spread of the

infection. Each of these elements contributed to increased costs and lower

efficiency in the transportation of finished goods. (Zhu, et al., 2020)

This impact is particularly significant because, as noted by Rahman et al. (2021,

p. 1), the shipping sector accounts for 90% of global trade and can be considered

the “artery of international supply chains”. Indeed, the authors also highlight

that, due to the pandemic, the volume of cargo transported dropped by 13% in

mid-April 2020, and forecasts announced a 32% reduction in the following

months.

Forecasts produced in February 2020 anticipated a reduction in global shipping

at U.S. ports of 12.9% year over year in February 2020 and of 9.5% in March

2020 (Lierow, et al., 2020).

Additionally, a survey conducted by the ISM (2020) highlights that during the

spring of 2020 62% of participants reported delays in receiving materials from

China, 48% were facing issues when moving goods inside the Chinese borders

and 46% experienced slower loading operations in Chinese ports.

An industry particularly affected by the decrease in passenger flights is the

pharmaceutical industry. As reported by The Washington Post Sunday (Duncan,

2020), by March 2020 the cost for airfreight increased from a few dollars per

kilogram to $15, even though the shipping rates for pharmaceuticals did not

increase disproportionately when compared to other cargo. This event

highlighted the high dependency of the USA’s pharmaceutical companies on

foreign drugs, particularly from China and India, especially after their request

to the Food and Drug Administration to prompt airlines to prioritize medical

supplies and international flights that transport them.

According to van Hoek (2020), globalization caused the lengthening of the

logistic pipeline, introducing additional risks of delays in the delivery process

and increasing dependency on remote sources. This mechanism undoubtedly

concurred in aggravating the impacts of the Covid-19 pandemic.

2.2.5 Relationship management

- Reduced social interaction

- Information ambiguity

- Lack of supplier engagement/opportunistic behavior

The sudden obstacle to smooth communication between actors involved in a

supply chain resulted in the incompleteness of the information shared,

increasing the ambiguity (already present even under normal circumstances due

to intrinsic complexities) under which decisions had to be made. Gunessee and

Subramanian (2020, p. 4) summarize several declinations of this phenomenon

based on the research available in operational management literature:

- Performance ambiguity, which leads to difficulties in evaluating the

efficiency of operations and therefore in deciding how to allocate

resources.

- Information ambiguity, inducing issues in interpreting the data available

and make decisions based on that.

- Causal ambiguity, strategically very important, which concerns the

uncertainty on the connections between outcomes and which events

caused them. “[It] could manifest itself in the supply chain context where

an organization is unable to determine how it has achieved a competitive

advantage as a result of some purchasing activity, or it could be a lack of

understanding of the linkages between inputs and outputs as related to a

supplier’s knowledge.” (Gunessee & Subramanian, 2020, p. 5)

- Extreme ambiguity, related to the lack of awareness regarding possible

future scenarios.

- Role ambiguity, meaning uncertainty on what is an actor’s position in the

supply chain hierarchy, which are their powers, and which are their

obligations.

- Relational ambiguity, which leads to a lack of understanding of how two

businesses are interconnected

- Processing time ambiguity, leading to incapacity to accurately plan.

- Uncertainty about probabilities, which might lead decision-makers to

gravitate towards choices with more certain outcomes and avoid

ambiguous ones.

Studies have shown that “decisionmakers benefit from trust, effective

communication and information exchange, and close relationships. […]

Strategically formulated social and environmental practices that are based on

long-term relationships and commitments – rather than mere tick-box

compliance exercises – can significantly increase organizational resilience”

(Trautrims, et al., 2020, p. 5).

GS1 US (2021) defines three main issues caused by a lack of visibility and

traceability inside supply chains which creates information ambiguity and

makes companies vulnerable to opportunistic behavior:

- Failure to recognize signs that indicated an imminent shift in demand

patterns

- Inability to respond to these shifts efficiently and effectively, for example

with changes in production. This inability could be caused by shortages

of raw materials or external constraints like new restrictions.

- Inadequacy in reallocating goods towards locations in higher need of

them.

A factor contributing to this issue is the lack of a monitoring system, currently

suffered by 44% of companies, which would allow them to address the

fluctuations in demand, supply, as well as promptly address challenges that may

arise. “A solid traceability system contributes confidence and trust to

organizations in the supply chain. Traceability helps ensure that companies can

confirm exactly where their products are in their life cycle and reduce the risk of

bad actors taking advantage of supply chain blind spots.” (GS1 US, 2021, p. 5).

2.2.6 Supply chain-wide impact (causing impacts in internal,

upstream, and downstream operations)

- Ripple effect on all the operations involved in supply chains

- Supply chain collapse

- Closure of facilities, including both companies’ production facilities and the

facilities of supply chain partners such as suppliers and distributors

Depending on a single source for any level of operation, from sourcing raw

materials to human labor, constitutes a great risk of disruption for supply

chains. When countries started to go on lockdown at the beginning of March

2020, the world’s largest 1,000 companies were deeply affected, as highlighted

in a study by the risk management company Resilinc (Linton & Vakil, 2020) and

visible in the figure below, as they possessed more than 12,000 factories,

warehouses, and other operations located in areas where activities where almost

entirely stopped to prevent the spread of infection.

Figure 8 - Dependence on quarantined areas (Linton & Vakil, 2020)

As reported by the Harvard Business Review (Haren & Simchi-Levi, 2020),

“mounting pressure to reduce supply chain costs motivated companies to pursue

strategies such as lean manufacturing, offshoring, and outsourcing. Such cost-

cutting measures mean that when there is a supply-chain disruption,

manufacturing will stop quickly because of a lack of parts”. Indeed, following the

efforts of the Chinese government to contain the transmission by quarantining

almost half of the population, many companies that relied solely on Chinese

manufacturers for sourcing components had to suspend production in some

plants due to the inability to find parts. As notable examples, Fiat Chrysler

Automobiles and Hyundai are mentioned.

Often, however, this issue proves particularly insidious when the disruption

involves second-level suppliers upstream, not the suppliers the company has

direct contact with, giving firms a false sense of security. This concern can be

explained using as an example the medical shortage faced by North America.

Long before the threat of the Covid-19 pandemic, the US Food and Drug

Administration carried out routine checks on their first-tier equipment

providers to assess the risk of shortages, finding them acceptable. However, they

failed to realize that those providers were sourcing raw materials from a narrow

portfolio of suppliers located in limited regions of India and China, pinpointing

the problem in an area of the supply chain that was further upstream than what

the FDA was focusing on. (Zhu, et al., 2020) As additional confirmation, a study

by the Institute for Supply Management found that the suppliers of 600 US

companies were operating at 50% capacity, leading to delays in the delivery of

final products (Ivanov & Das, 2020).

This issue represented a weak link in the chain and consequently a great risk.

Indeed, “disrupted global supply chains have had the biggest impact on

operations and the global economy in general” (Trautrims, et al., 2020, p. 3).

2.2.7 Financial management

- Reduced supply chain financial performance (e.g., loss/reduction of

financial stability)

- Reduced cash inflow

As a consequence of the negative impacts on production, during February 2020

companies on average were forced to reduce their revenue targets by 5.6%, with

the hardest-hit companies going down as far as 15% (ISM, 2020). Just one

month later, the figures significantly worsened reaching 22% (ISM World,

2020).

Some industries, like airline transport, have seen such considerable losses to

have forced companies to ask for governmental bailouts in order to keep the

business running and avoid leaving thousands of workers without employment

(Hakovirta & Denuwara, 2020).

The reduction of liquidity has been particularly challenging for SMEs. At the

beginning of the pandemic, 60% of Chinese SMEs reported availability of cash

sufficient for only two months’ worth of fixed costs. Similarly, in the USA small

business have on average enough cash flow to last only 27 days. This factor

constitutes a great vulnerability for SMEs, placing many at risk of permanent

closure. (Albaz, et al., 2020)

2.2.8 Sustainability management

- Lack of focus on social and environmental sustainability

practices/disruption of sustainability initiatives

- Threats to the health and safety of the workforce

- Contraction of the development of green and low-carbon energy sources

- Increase in waste

- Increase in recyclable materials

The Covid-19 crisis has also impacted the attitude of firms towards

sustainability, both economic and environmental.

The pandemic brought companies to increase their focus on “employee health

and wellbeing, helping employees to cope up with remote working conditions

[and] skill improvements.” (Sharmaa, et al., 2020, pp. 4-5). However, the

presence of other studies stating that Coronavirus may have worsened the

likelihood of exploitative conditions for workers in prone environments

(Trautrims, et al., 2020) raises the question of whether the previously

mentioned focus is just of performative nature, in order to improve the public

image of companies in these uncertain times, or if there are indeed two opposite

trends happening.

A study by the Economic Policy Institute (GS1 US, 2021) attests that 55 million

essential workers employed in 12 different industries are not equipped with

adequate protective devices, despite engaging in activities like food preparation

or tasks that are heavily based on human contact like retail customer services.

However, according to the RSM (Goel, 2021), “70% or more of manufacturers

took measures like asking sick workers to stay home, increasing the cleaning of

workplace surfaces, and encouraging workers to avoid face-to-face meetings”.

Additionally, many projects aimed at advancing green initiatives, like

implementing renewable energy sources, will be suspended because of a lack of

funding and more pressing priorities. For example, Morgan Stanley planned to

decrease the installation of US solar photovoltaics by 48%, 28%, and 17%

respectively in the last three quarters of 2020. Furthermore, the disruptions in

supply chains have affected the production of components necessary for these

implementations: “Many parts for large-scale renewable projects come entirely

or partially from China, other parts of Asia, or the United States. These are

specialized supply chains with few ready substitutes. The COVID-19 outbreak

has already slowed Chinese production of solar panels and materials, delaying

projects in countries including India and Australia. Manufacturing disruptions

in China could contribute to a significant one- or two-year dip in renewable

additions.” (Fox-Penner, 2020)

Regarding waste, the sudden surge in the use of single-use items - like surgical

masks, gloves, hospital supplies, etc., - has undoubtedly increased the pressure

on waste management operations. Similarly, those industries characterized by

perishable items which saw a drop in demand had to face the degradation of

their material stock and its subsequent disposal. (Dente & Hashimoto, 2020)

3 Building resilience

The current chapter will offer insights on why increasing the resilience of

companies is a fundamental practice and how that can be done, by firstly

formalizing the process of risk management and later investigating some

strategies that can be implemented to increase the resilience of companies.

According to the same study by the Institute for Supply Management mentioned

previously (ISM, 2020), as of the beginning of March 2020 almost 45% of the

companies surveyed had not prepared a resilience plan in case of disruptions in

their supply chain. Among these, 23% have reported disruptions in their

operations as of spring 2020. However, in a survey conducted in May 2020, 93%

of the supply chain executives interviewed declared intention to increase

resilience through concrete plans, with 44% of those being willing to do so even

at the expense of short-term savings (Lund, et al., 2020).

“Given the increasing frequency and intensity of natural disasters as well as the

continuous stream of anthropogenic catastrophes, the riskiest thing a company

can do is to have no contingency plan” (Fahimnia & Jabbarzadeh, 2016, p. 307).

According to experts at McKinsey Global Institute (Lund, et al., 2020) supply

chain disruptions that last more than a month can now be expected to happen

on average every 3.7 years, implying major consequences for companies.

Additionally, 80% of global trade involves countries with declining political

stability, increasing the risk of shocks to supply chains. Based on probabilities,

companies can expect these shocks to cause on average a loss of 45% of one

year’s EBITDA over the course of a decade.

Figure 9 - Magnitude, lead time, and frequency of disruptions (Lund, et al., 2020)

All of the elements mentioned above serve as evidence to prove the importance

of adopting strategies to increase business resilience.

Additionally, it’s useful to mention that McKinsey and Company (Alicke &

Strigel, 2020) highlights the different instances in which either proactive or

reactive responses to supply chain risks are more appropriate, depending on the

type of disruptions at hand. Therefore, the authors indicate reactive strategies

as most appropriate for risks that are hard to anticipate, in order to respond to

the disruption after this has happened. On the other hand, if the ability to

anticipate the risk is fair, it is worth adopting a more proactive approach.

It should be noted that the graph in Figures 9 and 10 present the same measure

in the axes, but these are reversed.

Figure 10 - Most appropriate set of response strategies depending on the type of disruption (Alicke & Strigel,

2020)

3.1 Risk management

Since section 3.2 will be dedicated to reviewing practical resilience strategies,

which are none other than applications of risk management to avoid or decrease

future repercussions should a disruption like Covid-19 happen again, it would

be helpful to firstly formalize what exactly risk management is and what process

it follows.

According to van Hoek (2020, p. 2) “existing supply chain resilience literature

would categorize panic buying as a demand risk, and the closure of supplying

factories and warehouses as a typical supply risk”. It is clear that risk exists at

various levels of the supply chain, and risk assessment is subject to the opinion

of those who are assessing since each individual will have his or her own opinion

about what may constitute a risk.

Barbara Gaudenzi and Antonio Borghesi (2006, p. 1) quote professor Martin

Christopher when defining supply chain risk as “any risk to the information,

material and product flow from original suppliers to the delivery of the final

product”. Generic risk management should aim to protect companies by

identifying unfavorable situations that could constitute a risk and then stopping

the negative events or reduce the consequences that could be detrimental for a

business while helping the recovery process after the crisis has passed (Slack, et

al., 2016). Risk management applied to supply chains, on the other hand, is

considered a supporting process to aid the achievement of predefined goals.

Risk management consists of four main activities, explained more broadly in the

next sections:

- Assessment of potential failures

- Prevention

- Mitigation, which is minimizing the negative consequences

- Recovering from failures when they do occur

3.1.1 Assessment

The first and most critical phase in the risk management process is to assess

activities in order to find potential sources of risk. Often this stage is what

determines the severity of the consequences since failure to detect a low risk

could prove to be more detrimental than a great risk that the company prepared

for. “Whatever approach to risk is taken, it can only be effective if the

organizational culture that it is set in fully supports a ‘risk-aware’ attitude”

(Slack, et al., 2016, p. 619). In the relevant literature, possible events with

negative consequences are defined as “failures”. Slack et al. (2016) provide a

checklist of potential causes of failure to analyze when beginning the risk

management process:

- failures of supply: fast progress and changes in products, together with

market fragmentation, have determined an elastic demand. Additionally,

the shift towards lean inventories and supply chain efficiency, as

discussed before, has caused companies to be highly dependent on their

outsourced activities.

- internal failures such as those deriving from human, organizational and

technological sources: these can be determined by human mistakes,

either voluntary or involuntary, poorly designed organizational structure

or processes, or faults in the facilities caused by lack of maintenance or

external undermining.

- failures deriving from the design of products and services: often

companies are pressured to meet a fast time-to-market performance at

the expense of accuracy in the design process.

- failures deriving from customer failures: client misuse could cause the

performance of products to be perceived inaccurately, therefore

companies should take on the responsibility to educate customers and

provide easy-to-use products.

- general environmental (or institutional) failures: this category includes

political disruptions and natural disasters. Unquestionably, Covid-19

falls in this category. Slack et al. (2016, p. 622) argue that “this source of

potential failure has risen to near the top of many firms’ agenda due to a

series of major events over recent years. As operations become

increasingly integrated (and increasingly dependent on integrated

technologies such as information technologies), businesses are more

aware of the critical events and malfunctions that have the potential to

interrupt normal business activity and even stop the entire company”.

However, sometimes sources of failure are difficult to identify. For this reason,

it is valuable to analyze previous disruptions and their root causes to build a

learning set for future reference. This post-failure analysis includes the following

activities:

- accident investigation: an examination of large-scale events carried out

by experts, which is important to carry out accurately since there is a low

number of cases to analyze due to their low frequency.

- failure traceability: the act of ensuring that all failures can be traced and

linked to the processes they went through. This helps recall entire batches

of products or analyze the chain that produced and handled a certain good

to find the element at fault.

- complaint analysis: gaining feedback from customer complaints in order

to find faults and how they are perceived by final consumers.

- fault-tree analysis: a logic map built starting from a failure to examine

possible causes, other consequences not yet detected, and ways to

improve goods and services.

After determining possible causes of risk, managers should estimate the

likelihood of a failure occurring. These estimates can be objective or subjective.

(Slack, et al., 2016).

Objective estimates are computed using historical performance based on data

collected, they can be measured using failure rates – how often a failure occurs;

reliability – the chances of a failure occurring; or availability – the amount of

available useful operating time left after taking account of failures. Subjective

estimates are more convoluted and unreliable since they are based solely on the

judgment of the individual (or team) making the estimate, who is not perfectly

rational by nature and who could have a different attitude towards risk than the

company. When objective estimates are unattainable, subjective estimates,

although not ideal, are preferable to no estimates at all.

The next stage is to assign priorities to risks in order to decide which to tackle

first with preventative and corrective actions. The approach described by Slack

et al. (2016) is failure mode and effect analysis (FMEA), used to “identify the

factors that are critical to various types of failure as a means of identifying

failures before they happen”. It assigns a risk priority number calculated based

on the answers to three key questions:

- What is the likelihood that failure will occur?

- What would the consequence of the failure be?

- How likely is such a failure to be detected before it affects the customer?

A very similar approach for assessing risk priorities by assigning numerical

values, the “Probability-Impact Matrix”, is explained by Dumbravă & Iacob

(2013). Often adopted in project management and emergency management, it

consists in evaluating risks based on two variables, the likelihood of it happening

and the impact it would have in case it happened. Many variations of this

method exist; however, the general process is building a matrix based on the two

variables, either with quantitative or qualitative values. In the case of

quantitative values, the two are multiplied to find the final risk score. Usually,

the matrix table is labeled with colors or letters to represent how critical the risks

are and therefore the priority with which managers should tackle those issues.

An example of the probability-impact matrix is provided in the figure below.

Figure 11 - Probability-impact matrix for risk assessment, with rating key (author’s elaboration, template by

Smartsheet)

3.1.2 Prevention

Once the causes of failures with the highest priority have been established,

prevention methods can be implemented to avoid the occurrence of negative

consequences.

Three main approaches are used (Slack, et al., 2016):

- Redundancy: having backup systems or components in case a process

or part of it fails. For example, companies that need to provide fast

services, like hospitals, have staff on call in case people on the current

shift have setbacks or more personnel are required.

- Fail-safeing: implementing systems or devices to prevent human

mistakes.

- Maintenance: taking care of facilities and resources to enhance safety,

increase reliability, provide higher quality, lower operating costs,

ensure a longer life span, and achieve higher end value.

3.1.3 Mitigation

When the occurrence of a shock cannot be avoided, therefore making prevention

methods not sufficient, mitigation can help to minimize the negative effects of

the adverse event. This step is essentially about minimizing exposure to shocks.

Several actions for mitigation exist (Slack, et al., 2016):

- Mitigation planning: ensuring that all possible outcomes have been met

with an action plan.

- Economic mitigation: taking on insurance or hedging against failures.

- Spatial containment: stopping the spread to other physical parts of the

facilities.

- Temporal containment: containing the spread of the effects of a failure

over time.

- Loss reduction: taking deliberate action to implement systems that do not

directly mitigate the failure but can reduce its negative effects.

- Substitution: comparable to the concept of redundancy mentioned

before, however, it does not always imply excess resources, it could

merely involve having backup plans in case the current one is deemed

inapplicable due to new circumstances.

An important element that aids in the mitigation process is the presence of an

emergency operations center (EOC). The presence of these departments is

usually limited to the corporate or business unit level, however, companies could

greatly benefit from their implementation to a deeper extent. These centers

could provide “predetermined action plans for communication and

coordination, designated roles for functional representatives, protocols for

communications and decision making, and emergency action plans that involve

customers and suppliers”. (Rice, 2020)

3.1.4 Recovery

After the failure has occurred, companies can engage in actions to recover and

benefit from the knowledge gained by the experience. The systematic approach

is aimed at “discovering what has happened to cause failure, acting to inform,

contain and follow up the consequences of failure, learning to find the root cause

of the failure and preventing it from taking place again, and planning to avoid

the failure occurring in the future.” (Slack, et al., 2016, p. 668).

In the subject matter of supply chains, and specifically their readjustment after

the impacts of the Covid-19 pandemic, the recovery process would necessitate a

revision of its structure. Changes in the flows and configurations could be

necessary in order to implement contingency plans developed with data

collected during the disruption. Additionally, collaboration between key actors,

both in terms of restorative actions and knowledge sharing, would be

fundamental elements to overcome the disruption. (Jabbour, et al., 2020)

3.2 Resilience strategies

Given the disruptions caused by Coronavirus that have been presented

previously, it would be constructive to review some strategies analyzed in

literature and applied by industry players that could help companies increase

their resilience and therefore mitigate adverse repercussions.

The instances illustrated in the following pages are practical applications of the

four steps of risk management described previously. Some sections deal with

assessment strategies, therefore only implying a passive analysis of the current

state of the supply chain to increase awareness and transparency. Others, on the

other hand, present actual changes in the organization of activities and in the

structure of the business, sparked by the know-how gained thanks to the current

sanitary crisis in order to prevent future reoccurrences.

3.2.1 Assessing the supplier network structure

The structure of the network of suppliers plays an important role in the

vulnerability of a supply chain and of the entire company.

For example, the interconnectivity of the network is an important factor to

consider. Actors that play a central role in the ecosystem could potentially

disrupt the network in a disproportional way. This is often the case of technology

providers who act as a de facto common utility for the network (Lund, et al.,

2020).

The ecosystem analysis should also include competitors within the industry as

well as correlated industries. If a supplier providing input to two different nodes

in the system suffers a breakdown, the company providing it will the least

business could be sacrificed in an attempt to minimize damage to relationships

with big buyers. Alternatively, a disruption limited to one industry could cause

difficulties to other industries that are dependent on the first one. For example,

Covid-19 caused ethanol production to decrease because of a drop in gasoline

sales. This issue affected the price of the CO2 used by companies producing

carbonated beverages, impacting their material costs. (Lund, et al., 2020)

This newfound need for transparency has increased the attention put on the

entire chain of suppliers, focusing beyond just a firm’s immediate supplier. The

episode of medical shortage in North America mentioned in section 2.2.6

emphasizes the need for greater scrutiny and transparency of the complete chain

of suppliers, with a proactive effort of information sharing, from initial sourcing

to finished products.

Indeed, Choi et al. (2020) testify that “companies that invested in mapping their

supply networks before the pandemic emerged better prepared” thanks to the

increased visibility into the structure of the chain, which allowed them to benefit

from the abundance of information readily available as soon as the disruption

happened. This data provides insights on the suppliers, locations, components,

and products most at risk, granting a head start in securing inventory and

switching production between sites.

A study by McKinsey (Lund, et al., 2020) found that tier-one suppliers are often

publicly disclosed, and companies manufacturing complex products, like

aircrafts, have the greatest number of tier-one suppliers. However, the study

found also that the entire network of suppliers expands exponentially beyond

those directly in contact with the company, with a total number that is 7 to 17

times the number of tier-one suppliers, and that more than 33% of disruptions

to supply chains occur beyond the second tier.

For instance, the JIT strategy (further described in section 3.2.4 “Increasing

stock”) causes the resilience of the company’s production to rely solely on the

resilience of its suppliers by minimizing the amount of stock. Therefore, a

comprehensive analysis of the reliability of the providers should be conducted.

This investigation could reveal that the companies able to grant continuity in

adverse conditions are also less convenient economically, or that no agent would

be able to withstand a disruption, maybe due to the nature of the industry itself,

exposing the company to risk.

The figure below features the characteristics of different structures of supplier

networks, highlighting in particular how they promote or hinder resilience.

Figure 12 - Impact on resilience and vulnerability of different supplier network structures (Lund, et al., 2020)

3.2.2 Stress testing

Following the financial crisis of 2008, a “bank stress test” was introduced to

determine a bank’s financial strength. This assessment consists in running

what-if scenarios to check whether an institution possesses sufficient assets to

endure times of economic stress. (Pritchard, 2021)

Much similarly, in order to address the risk created by low-probability high-

impact events like calamities or pandemics, some economists at Harvard

Business Review (Simchi-Levi & Simchi-Levi, 2020) developed a mathematical

model useful to understand the risks and impacts associated with disruptions

along a supply chain. The model pivots around two main concepts:

- Time to recover (TTR): “the time it would take for a particular node in the

supply chain — a supplier facility, a distribution center, or a

transportation hub — to be restored to full functionality after a

disruption” (Simchi-Levi, et al., 2014). The values identified by this

metric are based on historical data and on assessments of buyers and

suppliers and can differ across nodes of the supply chain.

- Time to survive (TTS): “the maximum duration that the supply chain can

match supply with demand after a facility disruption” (Simchi-Levi,

2015). This second metric was added afterwards due to the tendency of

suppliers to be too optimistic about the time needed to recover from an

interruption of their normal flow. Therefore, a measure to identify which

suppliers’ performance is more sensitive to accuracy in TTR disclosure

was needed.

The figure below is a graphical representation of this model as an example,

applied to the supply chain of the American automobile manufacturer Ford.

Figure 13 - Example of TTS metric applied to Ford's supply chain (Simchi-Levi, 2015)

The model aggregates the data from the two previously mentioned metrics with

any additional information available (bill-of-materials, relationships between

actors in the supply chain, inventory levels, etc.), offering a representation of the

network of dependencies happening in the supply chain.

Broadly speaking, when the TTS of a specific activity is lower than the

corresponding TTR, that node is likely to expose the firm to disruptions because

it will not be able to satisfy demand with supply for the entire time needed to

recover. Therefore, inventory can be strategically placed where it will cause the

TTS to be greater than the TTR “and thus a disrupted node will always recover

before it exceeds its ability to apply the mitigation strategies the firm has in

place”. (Simchi-Levi, 2015).

More accurately, this model runs scenarios simulating a disruption in one node

of the chain at a time, accounting for different levels of severity, determining

which type of response would minimize the impact of the disruption on the

performance of the firm.

Depending on the optimal response determined (reducing inventory, modifying

production, adjusting transportation, etc.) the model offers a financial or

operational performance impact (PI) for each node analyzed. This dimension

can be embodied by different measures of performance, like units or revenue.

The node in the chain presenting the highest loss in performance, and therefore

the highest PI, is paired with a risk exposure index (REI) of 1.0, while all the

other nodes are assigned values that are relatively computed from this first one,

making the overall scale ranging from 0 to 1. These indexed scores allow firms

to rapidly identify which activities in the supply chains are most at risk and

therefore require the most attention. (Simchi-Levi, et al., 2014)

3.2.3 Diversifying sourcing

Many companies rely on China for raw materials, components, and even

finished products, however during the year 2020 it has quickly come to attention

that if an entire country shuts down due to a pandemic, several industries are

forced to come to a halt. Diversifying to more than one supplier and choosing

suppliers from different countries can stem this problem. This is called plus-one

diversification, an approach that was already becoming common even before the

pandemic, due to the rising cost of Chinese labor and the resulting search for

more economically advantageous options. (Zhu, et al., 2020)

For instance, Foxconn, a contract manufacturer of Apple, has decided to start

shifting production towards India with an investment of $1 billion to diversify

its supply chain needs across multiple regions (GS1 US, 2021).

3.2.4 Increasing stock

One very common and straightforward way to avert a shortage of supply, and

therefore a disruption of business, is to increase stock. Many authors in

literature, among which Chiaramonti and Maniatis (2020), argue that

maintaining a sufficiently large strategic storage is critical to ensure continued

business.

The practice of keeping stock in a warehouse has been abandoned in the last

decades in favor of lean methodologies and the Just In Time strategy, “an

integrated set of activities designed to achieve high-volume production using

minimal inventories of parts that arrive exactly when they are needed” (Jacobs

& Chase, 2018, p. 18). This approach has many advantages, first and foremost

efficiency and cost saving, but it also leads companies to be highly reliant on

their supply chain, which is often very far from the company’s production site:

if the delivery of materials is delayed or there is a shortage, the factory’s activities

inevitably get halted (Zhu, et al., 2020). The inadequacy of these strategies stems

from the use of historical data to forecast production and stock needed, rarely

considering the chance of major disruptions like calamities or pandemics

happening (Simchi-Levi & Simchi-Levi, 2020).

Therefore, considering the consequences faced due to the Covid-19 pandemic,

the JIT approach should be re-evaluated in other to weigh the undeniable

positive aspects against the negative ones. Companies should analyze their

specific situation, given the industry they belong to and the portfolio of products

and services they offer, and question if higher cost efficiency due to no

warehousing is worth the loss in resilience. A good compromise could be

increasing (or establishing when not present) the buffer amount of stock in order

to grant the continuity of production even in the initial phases of an event like

the Covid-19 pandemic, where self-sufficiency is a crucial discriminant between

losing a high share of business or retaining performance while potentially

acquiring the market share lost by other less resourceful competitors.

Indeed, after the first wave of the pandemic, 47% of supply chain executives

declared intention to increase the amount of inventory for critical products, and

19.6% plan on keeping more inventory overall (Finances Online, 2021).

Section 1.4 argued good reasons for maintaining an inventory and how this

practice can benefit businesses. It also explained how the practice of

maintaining a sufficient amount of stock can increase the independence of

manufacturing operations from fluctuations in material availability.

3.2.5 Reshoring and domestic production

Even after the emergency has subsided, companies should focus on offshoring

towards selected countries near the main production site in China, in order to

foster easy communication and exchanges between factories while lowering

their dependency and therefore risk exposition. Additionally, by moving part of

the production to countries adjacent to China, other states will recognize the

opportunity to compete with Chinese suppliers. This would induce legislators to

produce policies that would be advantageous to companies attracted by low

costs and high diversification, convincing them to externalize inside their

territory. (Zhu, et al., 2020)

There would still be the negative aspect of uncertainty: since firms would be

investing in unfamiliar countries, they would face an initial learning curve

regarding the local culture and how to effectively manage relations with the new

branch factories. Even though the risk of the entire region facing a disruption

like a total lockdown still exists, the probability of it involving all the factories of

a company situated in different states is remote, and either way, it would be

impossible to avoid this risk – or any risk – all together. (Zhu, et al., 2020)

Moreover, several firms in the USA moved factories closer to their home

country, for example in Mexico, in order to have greater influence and control

on their supply. However, this may not always be possible, due to existing

binding contracts or difficulty finding similar quality and expertise elsewhere.

(Zhu, et al., 2020)

Another option consists in moving production in-house. Chiaramonti and

Maniatis (2020), for example, argue that maintaining the availability of

necessary supplies that are strategically fundamental heavily relies on the ability

to maintain some level of domestic production. Although localizing can be very

good for brand image by appealing to ethically conscious consumers that prefer

to shop locally or consumers attracted by products made in their home country,

and overall increasing the perceived value of products, it also implies much

higher costs, both of labor and of materials, requiring a budget that may simply

not be available, especially in the short term (Zhu, et al., 2020).

Indeed, a report by McKinsey (Lund, et al., 2020) estimates that between 16%

and 26% of global production for export could potentially move to a different

location in the future, shifting either to domestic production, nearshoring, or

offshoring to a new nation. The main industries involved in this estimate are

pharmaceuticals, petroleum, apparel, and communication equipment.

3.2.6 Technological advancements

The pandemic was also the catalyst for the increasing adoption of advanced

technology and automation already in motion in recent years. “Technology and

data can make supply chains more efficient, so they run more smoothly and

deliver greater value to customers, partners and the company.” (Goel, 2021).

Automation

Firstly, introducing more automated machines in factories significantly reduces

the need for human labor, since robots can work in unfavorable or unsafe

conditions, allowing production to continue even when safety measures

mandate fewer workers to be present in the workspace (Zhu, et al., 2020).

Additionally, robots are quicker and more effective in carrying out certain tasks,

especially when those tasks are dangerous for humans or very repetitive, for

instance in industries that are recently gaining more traction like material

sorting and upcycling (GS1 US, 2021).

For example, the British online grocery store Ocado, known for its e-commerce

and automated warehouse technology which are already being sold to other

supermarkets globally, has recently acquired two robotics companies from the

US, intending to eventually create a fully automated “dark” order fulfillment

center which requires no human presence (Kahn, 2020).

Internet of things

Employing new ways of monitoring processes, like the internet of things,

increases efficiency and the flow of information available to higher

management, favoring more informed decisions (Zhu, et al., 2020).

According to an article on Forbes (Williamson, 2020), a prediction formulated

before the beginning of the pandemic was forecasting more than 5.8 billion

endpoints in the Industrial Internet of Things (IIoT) ecosystem by the end of

2020. This technology can prove to be indispensable for companies to face future

challenges and disruptions like this pandemic. IIoT sensor networks can provide

manufacturers with the data necessary for fast decision-making.

Additionally, thanks to the data offered and the intelligence to analyze it in a

valuable way, it can show companies where to concentrate the resource that may

be subject to shortages at that moment. Moreover, inventory management

systems based on IIoT can potentially reduce inventory levels by as far as 36%.

Lastly, since the IIoT ecosystem provides real-time data on supply and especially

demand, it offers valuable insights and predictive analytics that are helpful to

avoid ripples of the bullwhip effect to cause drawbacks in supply chains.

(Williamson, 2020)

Digital twin

One factor that can help companies achieve higher resiliency against disruptions

is increasing the visibility during the entire life cycle of a product. This can be

done by applying a “persistent identity” to materials, components, and products

through a technological tool called “the digital twin”, a technological tool strictly

intertwined with the internet of things (GS1 US, 2021).

According to Kritzinger et al. (2018, pp. 1016-1017), the digital twin is “a digital

informational construct about a physical system, created as an entity on its own

and linked with the physical system in question. [It] should optimally include all

information concerning the system asset that could be potentially obtained from

its thorough inspection in the real world”. This virtual representation is

characterized by synchronization between the digital and real version of the

system in question through the use of data collected by smart devices, and it’s

capable of running many simulations using mathematical models and advanced

data elaboration.

The authors offer three main areas in which the digital twin can help boost

productivity and increase competitiveness:

- Production planning and control: orders can be planned according to

statistical assumptions derived from data collected; decision-making can

be supported by a detailed diagnosis offered by software; automation of

plans and order placing.

- Maintenance: the impact of variations upstream or downstream can be

easily pinpointed and assessed to avoid unwanted consequences on

processes; preventive maintenance measures can be developed more

easily; machine learning algorithms can help with the assessment of the

conditions of machinery during the life cycle of products, allowing for

more transparency and efficiency of data sharing.

- Layout planning: planning of the production systems is made easier, as

well as the evaluation of the current status.

Additionally, three main benefits of applying this technology can be observed:

(GS1 US, 2021)

- Ensuring product confidence and authentication

- Augmenting the repair, recycle, and reuse process with comprehensive

product history

- Returning the authentic product’s materials into the manufacturing

process

The importance of the last two advantages mentioned has been analyzed

previously in section 3.2.7 “Increasing circularity”.

Online stores

Additionally, many companies invested in the creation or improvement of

online stores or the presence of their products in online marketplaces to retain

sales and appeal to a larger audience given the limitations to travel and mobility.

However, this change often requires specialized skills, knowledge, and

investments that may not be already available to companies, especially small

ones, preventing them from reaping the benefits of this evolution. (Zhu, et al.,

2020)

Indeed, according to statistics studies (Finances Online, 2021), during the

pandemic “64% of retailers were challenged to adapt their supply chain for e-

commerce”.

3D printing

Furthermore, new technologies like 3D printing have allowed some companies

to quickly adapt production to be able to respond to unexpected changes in

demand for specific goods. This is the case of Naturepedic, a company

specialized in mattresses, who thanks to 3D printing quickly switched the use of

its cotton to produce face masks. (GS1 US, 2021)

3.2.7 Increasing circularity

“When faced with resource shortages, product scarcity, and limited traceability,

the supply chain became susceptible to counterfeit goods, compromised quality,

and delays in distribution.” (Nuce, 2021). Efforts to cope with these issues, and

therefore increase the resilience of the supply chain, have been found by a GS1

US (2021) whitepaper to be highly correlated to sustainability and a circular

economy.

By increasing the circularity of supply chains companies can increase the

productivity of infrastructures, products, and assets since their permanence

inside the chain is longer and therefore their derived value higher. This will lead

supply chain streams to benefit from remanufacturing new and existing

supplies, allowing emerging industries, like secondhand markets and waste

miners, to grow their size and importance. (GS1 US, 2021)

Closed-loop supply chains, characterized by the practice of handling returns and

upcycling them, were addressed in greater detail in section 1.3.

Figure 14 – Circular supply chain (GS1 US, 2021)

Additionally, consumers have shown appreciation for sustainable practices and

products, leading to increased brand loyalty and higher demand, ensuring fewer

fluctuations in case of disruptions when compared to products deemed more

replaceable and less valued by users. (Nuce, 2021)

3.2.8 Altering production

As stated previously, in the case of disruptions, consumption habits change

drastically. During an emergency, a rapid and short-term solution for companies

who wish to retain their market share might be to alter the span of their product

portfolio or the volume of production.

For example, in March, the Italian newspaper Il Sole 24 ORE (Carli, 2020)

reported that many companies, especially from the fashion industry, converted

their production chain to cover the demand for surgical face masks that

skyrocketed soon after the pandemic spread. This change requires a steep level

of flexibility that may be harder to obtain if employees receive highly specialized

and narrow training that is difficult to expand once the need for it arises.

On the contrary, providing cross-training to most of the organization as a

standard practice prepares the company for quick adaptation in time of need

and facilitates brainstorming and a multidisciplinary approach among

employees, which can be beneficial also outside the scope of a disruption to

promote innovative ideas. (Zhu, et al., 2020)

4 Recommending resilience strategies

through decision support systems

Since the previous chapter has discussed risk management and practical

resilience strategies, it would be helpful to study what scenarios could occur

depending on which and how many strategies a company decided to implement,

in order to suggest a way to identify the best course of action. Therefore, the

current chapter will be dedicated to describing an analytical method that

companies could use to determine the best choices to make in order to prevent

or mitigate the adverse effect of a disruption.

The basic method that will be described is able to provide maximum added value

when the data used is accurate and company-specific, thus requiring

unrestricted access to records and information that are hardly granted to

individuals outside the organization in order to protect confidential

information. The analysis here proposed, therefore, aims at presenting one

methodological approach that could aid decision-makers by providing strategic

suggestions and insights. Section 4.4 will offer additional observations on the

potential of this tool.

4.1 Decision trees

The method analyzed in the following sections is based on decision analysis and

in particular decision trees. This technique shows the elements of the decision

that has to be made in a clear and straightforward way that highlights the

evolution of the issue at hand through time.

This tool is composed of branches (represented by lines), which connect to each

other through nodes (represented mainly by squares or circles). Each node

symbolizes a point in time. If it is rendered through a square, it is called a

decision node and it denotes a time when a decision has to be made. The

branches spanning from it represent the options that can be chosen. If it is

rendered through a circle, on the other hand, it is called a chance node (or

probability node) and it embodies a time when the result of an uncertain

outcome becomes known. In this instance, the branches signify the outcome

possibilities and are often accompanied by the probability of that outcome

happening, which have to sum up to 1. The graph proceeds from left to right

indicating time passing, so branches spam from nodes that already happened

(on the left) to nodes that haven’t happened yet (on the right). (Winston &

Albright, 2019)

In this chapter, decision trees will be used as a Decision Support System. DDSs

are methodologies used to support the decision-making process using analytical

tools. They are usually not completely autonomous, meaning that they are not

intended to completely substitute the decision maker since they are based on a

blend of factual data, business-specific knowledge, and estimates (Burstein &

Holsapple, 2008). Therefore, basing on the data collected and the structure of

the tree, the system will suggest a decision path as the most advantageous.

4.2 The general model

Given a generic company called Alpha, the sequence of decisions and outcomes

regarding the implementation of three unspecified strategies (X, Y, and Z) can

be represented with the decision tree below.

The points in time when Alpha is confronted with a decision are represented by

squares, and the two possible outcomes of the choice – in this case implementing

or not implementing the strategy in question – are indicated with a green or

orange arrow respectively. Hence, company Alpha starts from choosing whether

to implement strategy X or not, then, regardless of the outcome of the first

decision, proceeds with the same process regarding strategy Y, and then

similarly for Z.

It’s important to note that for the purpose of this instance, the outcomes

resulting from the implementation of each strategy are independent from those

related to the implementation of others, meaning that the decision to carry out

one plan will not influence the outcome following the implementation of the

others. If, for example, the choice to implement or not strategy X was taken after

the choice regarding strategy Y, the resulting situations would still cover all the

possible combinations. Additionally, the implementation of strategy X does not

affect the amount of mitigation arising from the implementation of another

strategy or the cost that it entails, and vice versa.

The decision tree with the relative key will now be presented below.

Figure 15 – Complete decision tree (author’s elaboration)

A 1

B 1

C 1

D 1

E 1

F 1

G 1

H 1

strategy X

strategy Y

strategy Z

disruption

Figure 16 - Key for the decision tree (author's elaboration)

All the possible combinations of choices lead to eight different situations,

represented on the right portion of the decision tree with circles and indicated

by the letters A through H. They are additionally outlined for clarity in the table

below, along with the combination of strategies that are implemented in each

one. Therefore, for example, situation F represents the choice to implement only

strategy Y but not X or Z. The last column, “”, represents the set populated by

the strategies that have been implemented in that situation and it will be used

in formulas in the following pages.

0 no disruption

1 small disruption

2 intermediate disruption

3 large disruption

decision node

chance node

-strategy implemented

-strategy NOT implemented

KEY

Table 2 - Combinations of strategies implemented depending on the situation (author's elaboration)

The eight situations coincide with chance nodes. The possibilities after these

points are determined by probability. For this model, four outcomes have been

chosen as a demonstration, and they consist of four variations of a possible

disruption. The first one, 0, identifies the chance of no disruption happening (or

a disruption that doesn’t affect the company in question in any way), while 1, 2,

and 3 represent three intensities of disruption, from weakest to strongest. The

same structure of decision tree and similar results would work also considering

one disruption, like the Covid-19 pandemic, and the different intensity of

impacts that it could have on a company, with the relative likelihood of that

impact being sustained.

SITUATION X Y Z

σN

A yes yes yes X, Y, Z

B yes yes no X, Y

C yes no yes X, Z

D yes no no X

Eno yes yes Y, Z

Fno yes no Y

Gno no yes Z

Hno no no Ø

STRATEGIES IMPLEMENTED

Figure 17 - Close-up of a chance node of the decision tree (author's elaboration)

The illustration will now proceed considering some simplified data, presented

in the table below. For the purposes of this model, the revenue will be expressed

as a value of 100 with no unit of measure in order to serve as the sole reference

for all the values of impact that will be computed later.

Table 3 - Data about the company, possible disruption, and resilience strategies (author's elaboration)

Two measures are indicated for each degree of disruption. The first one,

probability, indicates the likelihood of the disruption happening. These values

are complex to derive due to the nature of the problem. Many factors come into

Revenue 100

Disruption Probability Impact

0 0,17 0%

1 0,36 6%

2 0,27 20%

3 0,20 51%

Strategy Cost Mitigation

X 1,30% 45%

Y 0,81% 40%

Z 1,44% 50%

DATA

play and many iterations have to be considered when trying to compute the

likelihood of a supply chain disruption happening, thus explaining the difficulty

of finding precise figures in literature. However, an approximation has been

obtained from the study presented by Lund et al. (2020), visible in Figure 9 at

the beginning of Chapter 3. The data considered for the current analysis

corresponds to the three worse durations of disruption illustrated in the

aforementioned study, given that it seems more constructive to consider more

severe scenarios for the purposes of this study, along with being numerically

more appropriate. According to the report, a disruption lasting two to four weeks

can be expected to happen every 2.8 years; one with a medium duration of one

to two months occurs every 3.7 years, while a prolonged disruption lasting more

than two months can be expected every 4.9 years. These measures have been

used as a reference for the disruptions of severity 1, 2, and 3 respectively. From

these figures, the probability of occurrence can be computed, dividing 1 by the

frequency. The probability of no disruption happening, corresponding to type 0,

has been computed by subtracting the three probabilities from 1 to obtain the

residual amount, so that all possible outcomes of the chance nodes sum up to 1,

as the rules for decision trees require. The table below offers a visual account of

this process. Even though this way of deriving probability from frequency is an

extremely rough estimate, the high complexity of computing the likelihood of

disruptions and the lack of clear figures in literature has required such an

approximation.

Table 4 - Calculation of the probability of occurrence (author's elaboration)

The second value related to each degree of disruption, impact, represents the

negative consequences that the company would have to suffer in case the

disruption happened, and it’s expressed as a percentage of revenue, to keep the

measure relative. Unfortunately, a single source providing all the necessary

estimates for this figure is nearly untraceable, therefore the specific values

related to impact originate from different sources. An article relating to the effect

of a six-month disruption of the retail sector (Binlot, 2020) and one related to

the impact on small businesses (Arora, 2020) were used as a reference for the

impact of a very heavy disruption, type 3. The remaining data for types 1 and 2

has been derived from a report by GEP (Bartels, 2021), while, reasonably, in

case of no disruption the negative impacts would be none.

Regarding strategies, on the other hand, the two data presented are cost and

mitigation. The first one is expressed as a percentage of revenue, and it

represents the cost that the company has to incur to implement the strategy in

question. The second one, mitigation, is the positive effect given by the

implementation of the strategy, specifically the percentage of reduction of the

negative effect caused by the disruption. In the next section, the sources for these

two values will be explained in a case study using real data.

Disruption

Frequency

(years)

Calculation Probability

12,8 1/2,8 0,36

23,7 1/3,7 0,27

34,9 1/4,9 0,20

0 - 1-(0,36+0,27+0,20) 0,17

All the information presented can now be used to compute the total impact of

the disruption on the company, provided visually in the table below.

Table 5 - Overall impact of the disruption differentiated by degree of severity and situations considered, after

accounting for the mitigation effect of the strategies implemented (author's elaboration)

Since, as already mentioned, the revenue has been assigned a value of 100, all

the values of the overall impact can be considered a cost expressed as a percent

of revenue since they are normalized to it. Even with this reasoning, however,

the choice of omitting percentage signs for these figures has been motivated by

the pursuit of visual clarity.

To compute the values of the overall impact on the company visible in Table 5,

the cost of all the strategies implemented has been summed to the negative

impact determined by the specific degree of disruption corresponding to that

cell. The following formula better explains the process. The key with the

interpretation of all the notations can be found below.

0 1 2 3 AVERAGE

A3,6 4,5 6,9 12,0 6,51

B2,1 4,1 8,7 18,9 8,04

C2,7 4,4 8,2 16,8 7,68

D1,3 4,6 12,3 29,4 11,18

E2,3 4,1 8,3 17,6 7,64

F0,8 4,4 12,8 31,4 11,58

G1,4 4,4 11,4 26,9 10,42

H0,0 6,0 20,0 51,0 17,96

OV ERALL IMPACT

   󰇛 󰇜



󰇛 󰇜 󰇛 󰇜



Equation 1 - Formula for computing the total impact of a given situation N for a degree of disruption D (author's

elaboration)

Therefore, following the formula above, these steps are performed to compute

the total impact T for a determined situation N and degree of disruption D.

Firstly, the percentage cost of a strategy S is multiplied by the revenue, to reach

the actual cost of the strategy. This product is done for every strategy that

belongs to  and is, therefore, an implemented strategy. All the products are

then summed together, and the total value obtained up until now is the total cost

of implementing the strategies.

Subsequently, the revenue is multiplied by the percentage impact of the degree

of disruption in question, to attain the cost (or loss of revenue) suffered by the

company. This measure, however, gets multiplied by another factor to account

for the mitigation effect of the strategies implemented, which therefore lessens

the total damage inflicted by the disruption. This factor is computed by

multiplying together 󰇛󰇜 for every strategy that is implemented in the

current situation. This measure inside brackets signifies the amount of damage

remaining after the mitigation, therefore the product provides the overall

percentage of damage that is actually inflicted on the company after accounting

for the mitigation of the strategies implemented.

The last column of Table 5, on the other hand, is a weighted average and

corresponds to the value of overall impact that has been obtained by computing

the matrix product of the values of the first four columns by the probabilities

indicated in Table 3, as the formula below demonstrates. This provides a

comprehensive view of the risk incurred by the company.

 󰇛 󰇜



Equation 2 - Formula for computing the average effect of disruptions given the probability of them happening

(author's elaboration)

Since, as demonstrated in the previous paragraphs, the values in the impact

table are the sum of costs, the most convenient situation will be the one that

presents the lowest value. Following this reasoning, color-coding has been

added to Table 5, with shades ranging from dark green for low values to yellow

for high ones, to provide an immediate and intuitive interpretation of the

figures. The formula below translates this concept into mathematical language:

it returns the situation N that minimizes the previously calculated average

impact. It corresponds to the situation that implies the lowest cost to the firm

and, therefore, to the most advantageous.

 



Equation 3 - Formula for minimizing the average impact (author's elaboration)

In this instance, the recommended situation overall is implementing all three

strategies. As expected, in the first column of Table 5, where the degree of

disruption considered is 0, the suggested action is not proceeding with any

strategy, since it would be wasteful to invest in mitigation if no disruption

happened. However, as the degrees of disruption get more severe, therefore

moving towards the right in the table, the model advises more and more strongly

to implement an increasing number of mitigation plans. Given the percentage of

impact of the disruptions considered, this latter trend prevails, resulting in

situation A to be suggested based on the average impact column.

Below, a table with the meaning of all the notations used in the formulas is

presented. The numeric values of the notations used can be found in the figures

and tables situated in the previous pages.

Table 6 - Key of notations used in formulas (author's elaboration)

Notation

Meaning



Overall impact for a specific situation N and a degree of

disruption D



Situation. Values: A through H



Degree of disruption. Values: 0, 1, 2, 3



Cost of a strategy S, expressed as a percentage of revenue



Strategy. Values: X, Y, Z



Revenue



Set of strategies that have been implemented in a situation N



Impact of a degree of disruption D, expressed as a percentage of

revenue



Mitigation given by a strategy S, expressed as a percentage of

reduction of the negative effect caused by the disruption



Average impact of the various degrees of disruption for a

situation N



Probability of a disruption D happening

4.3 The model considering synergies

Let us now consider a case, following the same decision tree, where the three

strategies are not generic, but specified. In particular, the following have been

chosen for this example.

The first strategy consists in increasing stock, as explained in more detail in

section 3.2.4. This strategy will be marked in the tables below as “stock”, and it

will replace strategy X on the decision tree. The value related to its cost has been

obtained from the financial data related to the Italian company Carel SPA, found

in the Orbis database (Bureau van Dijk, 2021). The company specializes in

control solutions for air-conditioning in the electrical and electronic

manufacturing sector, therefore the choice of examining it for this study has

been motivated by their heavy reliance on raw materials, which makes them the

ideal candidate to identify the cost related to increasing the amount of stock to

avoid shortages in the supply of materials.

In 2019, Carel had an inventory amount of €54 million, which rose to €64

million in 2020, corresponding to an increase of 15.6%. The revenue from the

same years is €370 million and €400 million respectively, indicating a 7.5%

increase. Since the value of the inventory grew more than what would be

coherent with the growth in revenues, we can assume that the remaining

inventory of 8.1%, which was not used for production, lingered as emergency

stock. This amount corresponds to €5.2 million, which equates to 1.3% of the

yearly revenues. A summary of this data can be found in the table below.

Table 7 - Data on inventory and revenue for Carel SPA for 2019 and 2020 (Orbis database)

The second strategy is diversifying sourcing, addressed in section 3.2.3. It will

be referred to as “divers.” in the tables below and will take the place of strategy

Y on the decision tree. The cost relating to this strategy was once again obtained

by analyzing the data of the Italian company Carel, from the Orbis database

(Bureau van Dijk, 2021). In 2018, Carel SPA had costs amounting to €130

million for raw materials. By diversifying sourcing, the company would not be

able to benefit from some economic advantages they received from their primary

supplier, like bulk discounts and lower prices reserved to long-time trade

partners. We could therefore assume a yearly increase of 2% in the cost of raw

materials due to this, amounting to €2.6 million and 0.81% of the yearly

revenues, which can be used as a reference for the cost of this strategy. Table 8

offers a clearer understanding of this data.

2019

(mil. €)

2020

(mil. €) Δ%

Revenue 370 400 +7,5%

Inventory 54 64 +15,6%

%Mil. € % of Rev.

Emergency

stock

8,1% 5,2 1,30%

CAREL SPA

Table 8 - Data on revenue and cost of raw materials for Carel SPA for 2018 (Orbis database)

The third strategy is introducing automation as described in section 3.2.6, which

was dedicated to discussing technological advancements. This strategy will be

titled “autom.” and it will substitute strategy Z on the decision tree. The cost

related to its implementation has been obtained from the case of the online

grocer Ocado, already mentioned in section 3.2.6 and illustrated in the article

by Kahn (2020). The firm acquired in 2020 two U.S. robotics companies, for

$262 million and $25 million respectively, to increase the level of automation in

its warehouses. Since the value of this acquisition is of technological nature, we

can estimate that the investment will have an amortization period of 4 years,

which seems a reasonable measure for assets related to innovation. The article

also states that the acquisitions will increase the yearly revenues of the company

by $38 million, which would amount to $152 million over the period of

amortization. Therefore, we can consider the net investment in technology to be

the sum of the amounts paid for the acquisitions, minus the expected increase

in revenues over 4 years. This value, considered yearly and as a percentage of

revenue, equals 1.44%. Once again the following table clarifies the data used.

2018

(mil. €)

Revenue 320

2018

(mil. €)

+2%

(mil. €)

(% of Rev.)

Cost of raw

materials

130 132,6 2,6 0,81%

CAREL SPA

Table 9 - Financial data for Ocado (Kahn, 2020)

The peculiarity of choosing these strategies is that the first two, increasing stock

and diversifying sourcing, belong to the same impact area, supply management.

This implies that the mitigation resulting from their implementation is not

independently related: adopting one will influence the amount of mitigation of

the other. Therefore, the figures related to mitigation have been considered

differently for the two separate strategies and then for the case in which both

strategies are implemented at the same time.

Additionally, for this instance, the intensity of disruption has been considered

in terms of both duration and impact on revenue. Therefore, a disruption of

degree 1 will be a short interruption of business that causes small losses, while

#1 (mil. $) #2 (mil. $) Total (mil. $)

Acquisitions 262 25 287

Amortization

period (years)

Yearly

(mil. $)

Over amort.

(mil. $)

Exp. increase in

revenues

38 152

2019

(mil. $)

Exp. 2020

(mil. $)

Revenue 2300 2338

Total

(mil. $)

Yearly

(mil. $)

% of Rev.

(mil. $)

Net investment 135 33,75 1,44%

OCADO

one of degree 3 will be a prolonged disturbance with heavy losses. Consequently,

the amount of mitigation has now been considered variable depending on the

degree of disruption in question, since the efficacy of the strategies considered

varies depending on the length of the disruption that they have to mitigate.

In particular, increasing stock will be fairly effective at mitigating a short

disturbance, but this effect will sharply decrease as the time length of the

damage increases, since additional inventory would eventually run out and it

could be difficult to replenish due to prolonged supply interruption caused by

the negative event.

The same can be argued about the diversification of sourcing, despite presenting

a less sharp decrease of effectiveness: having a diversified supplier portfolio

would allow the company to not halt production completely due to lack of parts,

but it’s unlikely that during a prolonged disturbance suppliers would be able to

maintain an intensified production sufficient to compensate the loss of the

suppliers who were affected by the disruption.

Implementing both will not imply much of a premium during a short disruption

when compared with implementing only one of the two, since just one or the

other is enough to satisfy the supply needs over a short amount of time. As the

length of the disruption increases, however, having implemented both will pose

a great benefit, since their combined mitigation will allow a company to grant

the continuity of most of the business.

Introducing automation, on the other hand, would reasonably grant the same

mitigation effect no matter the duration of the disturbance, since its benefits

stem from two main reasons: firstly, automated machinery can adapt labor

efforts to the needs of the company; and secondly, it can continue production

when the disruption causes circumstances that would create problems for

businesses that rely on human labor, such as impediments to public

transportation that would hinder commuting or restrictions on assembly inside

factories.

Table 10 shows all the values that will be considered for this section. Besides

mitigation, which will be discussed below, the remaining data has been kept

equal to the data used for the general model.

Table 10 - Data about the company, possible disruption, and specific resilience strategies (author's elaboration)

Once costs for the strategies considered have been established, their mitigation

can be discussed. Table 11 presents the amounts of mitigation expressed as

percentages of reduction of the negative effect caused by the disruption, for the

single strategies plus the case of both increased stock and diversification

Revenue 100

Disruption Probability Impact

0 0,17 0%

1 0,36 6%

2 0,27 20%

3 0,20 51%

Strategy Cost

stock 1,30%

divers. 0,81%

autom. 1,44%

DATA

implemented at the same time, and the aggregate value for each combination of

situation and degree of disruption.

Unfortunately, the search for a reference for this value has yielded no

satisfactory results since the Covid-19 pandemic is still ongoing at the time of

writing and, even considering other disruptions that happened in the past,

finding a figure that could be universally adequate has proven to be challenging.

The amount of mitigation that the strategies considered can offer against the

degrees of disruption examined is extremely company-specific, therefore an

accurate value would be difficult to find without considering a determined

company and tailoring the entire analysis on it, asking managers and decision-

makers to provide precise data. For these reasons, the numeric amounts

indicated in the tables below consist of a reasonable estimate to serve as

exemplification, analogously to the ones offered for the general model. As

already mentioned previously, the focus of this analysis rests more on the

methodological approach examined rather than the specific data used, since

precise figures would need to be considered on a case-by-case basis.

Table 11 - Amount of mitigation accounting for synergies (author's elaboration)

The amounts in the aggregate mitigation table above have been derived by either

plainly reproducing the value from the upper table, which shows the mitigation

amounts for each strategy, or by combining two values of mitigation where

applicable. For example, situation A entails the adoption of all three strategies,

therefore the values used will be the one related to the row of “stock + divers.”

since both strategies are implemented and thus the value that considers

synergies has to be used, and the one related to “autom.”. These two measures

are aggregated using the formula below, where MT is the total mitigation

obtained and M1 and M2 are the two mitigation amounts to be combined.

STRATEGY 0123

stock 0% 60% 45% 30%

divers. 0% 45% 40% 35%

stock + divers. 0% 65% 75% 80%

autom. 0% 50% 50% 50%

SITUATION 0123

A0% 83% 88% 90%

B0% 65% 75% 80%

C0% 80% 73% 65%

D0% 60% 45% 30%

E0% 73% 70% 68%

F0% 45% 40% 35%

G0% 50% 50% 50%

H0% 0% 0% 0%

AGGREGATE MITIGATION

MITIGATION WITH SINERGIES

MITIGATION

 󰇛󰇜󰇛 󰇜

Equation 4 - Formula to aggregate the amount of mitigation for different strategies (author's elaboration)

The data presented above can now be used to compute the total impact of the

disruption on the company, presented in the table below, using the same

formulas presented previously for the general model in section 4.2.

Table 12 - Overall impact of the disruption differentiated by degree of severity and situations considered, after

accounting for the aggregate mitigation effect of the strategies implemented (author's elaboration)

Now an interpretation of the results obtained, analogous to the one seen in the

previous section, can be performed. As observed previously, the column

concerning the average impact in Table 12 highlights what the total costs

associated with each situation are. The most convenient combination of

strategies to implement, due to the formulas used, will be the situation

presenting the lower cost. Therefore, given the data employed, the overall most

convenient situation according to the model is A, i.e., implement all strategies.

As seen before, the number of implementations advised increases with the

impact of the disruption considered: column 0 suggests no strategies, 2 suggests

only a partial implementation, while 2 and 3 will suggest a complete adoption.

0 1 2 3 AVERAGE

A3,6 4,6 6,1 8,7 5,64

B2,1 4,2 7,1 12,3 6,29

C2,7 3,9 8,2 20,6 8,30

D1,3 3,7 12,3 37,0 12,42

E2,3 3,9 8,3 18,8 7,84

F0,8 4,1 12,8 34,0 12,00

G1,4 4,4 11,4 26,9 10,42

H0,0 6,0 20,0 51,0 17,96

OV ERALL IMPACT

4.4 Additional considerations

As stated at the beginning of this chapter, the method proposed aims at

providing an exemplification of how the implementation of resilience strategies

can be analytically considered to assess its feasibility and worthiness. Therefore,

it should be seen as a template to be customized by companies to fit their specific

needs.

For this reason, this section will now offer some considerations regarding the

proposed analysis, in order to provide reflections on the assumptions utilized,

insights into possible additional variables to factor in, and a more

comprehensive view on the potential applications of this study.

As previously explained, the overall impact is computed by summing the cost

related to each disruption multiplied by the chance of that disruption

happening. However, this approach could be considered the most appropriate

only for some companies. For example, firms focused mainly on short term

horizons, like start-ups, could prefer focusing on strategies that mitigate only

the most likely disruption, therefore choosing to ignore the risk posed by events

with lower probabilities which are likely to happen with a frequency that makes

them almost irrelevant to the short-term focus of the company. This means that

the situation most convenient in this case according to company managers

would probably not coincide with the situation derived by minimizing the last

column in the overall impact table, corresponding to the average values. In this

instance, the most suitable course of action could be, for example, considering

only the column of the disruption of type 1, being the one with the highest

probability and therefore most likely to happen in a short time horizon, and

finding the best situation among those values. Other companies, on the other

hand, could find it more appropriate for their strategical attitude to disregard

the column of the disruption of type 1 and base their decisions only on the

possible impacts caused by the more severe hazards. This could be the case if the

nature of the company or the industry in which it operates already intrinsically

implies a level of protection from the negative consequences of short or mild

incidents.

Another interesting variable that could be added to enhance and customize the

analysis is a study of the structure of the investments required to implement the

strategies. The formulas used in the previous sections didn’t differentiate

between a fixed initial cost or a deferred payment, they simply considered the

total expense related to that strategy. However, some companies might value

more positively strategies that require many small investments due to their lack

of liquidity, while others could prefer impacting the bottom line of the income

statement all at once on the year of implementation and then going back to the

ordinary level of profits in the following years.

Additionally, it could be insightful to take into consideration different degrees

of implementation of the strategies. In this case, the decision tree would not

present just two outcomes (implementing or not implementing) spanning from

the decision nodes, but possibly others depending on how many degrees of

implementation one wishes to consider. Options could be, for example, hard,

soft, or no implementation, contingent on variables like intensity or

exhaustiveness. In the case of increasing stock, this could mean by how much

the inventory is expanded. When taking into consideration technological

advancements, on the other hand, this could imply how much of the business or

supply chain is equipped with the new technology. Depending on the specifics

of the company considered, it could prove valuable to implement many

strategies to some degree to acquire a broad, even if incomplete, protection from

disruption. Alternatively, for businesses strictly intertwined in the network of

the industry they work in, the best option could prove to be implementing

thoroughly only sector-specific strategies and disregarding the others.

Additionally, managers could find it useful to enrich this study by performing a

sensitivity analysis on key variables to investigate the impact of their change on

the overall results.

For example, it could be useful to compute the minimum amount of mitigation

that each strategy would have to offer in order for the implementation of that

strategy to be advisable. Since this value would probably be difficult to measure

with a high degree of accuracy even having unrestricted access to company

records, this type of investigation could reveal how much margin of freedom the

company has. For instance, if the analysis revealed that by decreasing the

mitigation effects by just a few percentage points the implementation of that

strategy was no longer advisable, then the company could choose to not

implement it since there is a possibility of error in the input data. On the

contrary, if adoption was recommended even with much lower values of

mitigation, then managers could more confidently decide to proceed with that

strategy.

Similarly, it could be interesting to investigate the maximum costs of the

strategies that would still make their implementation the most convenient

situation. This investigation could be paired with the different degrees of

implementation of the strategies discussed in a previous paragraph in this

section. For instance, increasing stock might not be advisable to the extent

initially considered, since it implies too high of a cost. However, the sensitivity

analysis could reveal that if the required investment slightly decreased, then this

procedure would be advantageous. Therefore, the company could increase stock

by an amount lower than initially planned to fully reap the benefits of this

strategy. Additionally, if the analysis showed that a minor increase in the cost

would no longer advise implementation, then decision-makers could focus

efforts on ensuring that expenses stay within predicted amounts by, for example,

stipulating contracts with involved actors. This supposition is especially relevant

for instances with deferred payments or upkeep costs, where the likelihood of a

variation in costs is higher since part of them rests in the future.

Conclusions

People innately tend to underestimate the likelihood of negative events

happening, preferring to focus on the short term and ignore the distant future.

Empirical evidence of this is provided by the tendency of individuals involved in

business decisions to focus primarily on yearly or even quarterly results.

Managers, by receiving year-end bonuses depending on their performance

during this period, will be incentivized to concentrate on the short term for the

sake of their own personal gain. Similarly, shareholders are interested in the

returns they receive annually from the company's results. This means that both

of these categories will naturally be inclined to want to avoid investments that,

to protect against uncertain future events, will surely impact the company's

results in the short term.

This thesis provided an overview of the significant impacts that the Covid-19

pandemic has caused to economies and businesses around the world. In

particular, it demonstrated the high exposure to risks of supply chains by

analyzing the many different ways in which they can be affected by disruptions,

providing also practical examples and statistical data. This proved just how

pervasive the effects of a seemingly small negative event can be.

Additionally, it offered an account of how companies can improve their

resilience. This was done by firstly revising the ways in which risk management

has been theorized, therefore attesting to the importance of this process.

Secondly, it illustrated strategies gathered from literature and real-life cases

which can lower the risk exposure of firms, focusing mainly on tactics based on

prevention, requiring, therefore, a precautionary implementation that would

safeguard the company from possible negative events in the future.

Lastly, it demonstrated how analytical tools can support the decision-making

process and help understand which actions improve the resilience of businesses

the most. Particularly, it firstly offered a basic illustrative analysis to present the

method, using a sample approach based on decision trees. Later, it moved to a

more concrete study using real-life data, considering three strategies that were

described in the previous chapter. This way, the viability and usefulness of their

implementation were discussed, analyzing the benefits they offered against the

investment they required, to prove when their application was convenient for

the company. Lastly, by providing several additional considerations and

possible variables to add to the analysis, this thesis illustrated how these

instruments can be tailored to the needs and attributes of specific companies,

therefore proving their broad adaptability and significant value.

This study demonstrates that an increased focus must be put on the resilience of

companies and in particular of supply chains, for the gain of all economies and

businesses. Additional studies are required to clarify the features of possible

resilience strategies, in particular on the protection they offer against different

types of disruptions and on which companies could best benefit from them.

Hopefully, the events of the last two years will raise awareness on the

importance of being prepared for every occurrence and on how, in this deeply

interconnected world, one seemingly small occurrence can quickly spark a chain

of events leading to disruptions of global scope.

97 
References 
List of figures 
Figure 1 - Intermodal transport chain (The Geography of Transport Systems, 
n.d.) ....................................................................................................................... 4 
Figure  2  -  Total  integration  required  within  the  supply  chain  (Lummus  & 
Vokurka, 1999) ..................................................................................................... 7 
Figure  3  -  Simplified  graphical  representation  of  a  closed-loop  supply  chain 
(Fathollahi-Fard, et al., 2018) ............................................................................ 13 
Figure 4 - Benefits of decoupling points in reducing variability (Ptak & Smith, 
2016) ................................................................................................................... 16 
Figure 5 - Decoupled distribution network (Ptak & Smith, 2016) .................... 19 
Figure 6 - Extreme poverty in 2015-2021, measured as the number of people 
living on less than $1.90 per day (Mahler, et al., 2021) .................................... 23 
Figure 7 - Covid-19 impact on spending habits by segment (Ernst and Young, 
2020) ................................................................................................................... 29 
Figure 8 - Dependence on quarantined areas (Linton & Vakil, 2020) .............. 37 
Figure  9  -  Magnitude,  lead  time,  and  frequency  of  disruptions  (Lund,  et  al., 
2020) ................................................................................................................... 43 
Figure 10 - Most appropriate set of response strategies depending on the type of 
disruption (Alicke & Strigel, 2020) .................................................................... 44 
Figure  11  -  Probability-impact  matrix  for  risk  assessment,  with  rating  key 
(author’s elaboration, template by Smartsheet) ................................................ 49 
Figure 12 - Impact on resilience and vulnerability of different supplier network 
structures (Lund, et al., 2020) ............................................................................ 55 

Figure 13 - Example of TTS metric applied to Ford's supply chain (Simchi-Levi,

2015) ................................................................................................................... 57

Figure 14 – Circular supply chain (GS1 US, 2021) ........................................... 67

Figure 15 – Complete decision tree (author’s elaboration) ............................... 72

Figure 16 - Key for the decision tree (author's elaboration) ............................. 73

Figure 17 - Close-up of a chance node of the decision tree (author's elaboration)

............................................................................................................................. 75

99 
List of tables 
Table  1  -  List  of  impacts  of  the  COVID-19  pandemic  on  supply  chains 
(Chowdhury, et al., 2021) ................................................................................... 26 
Table 2 - Combinations of strategies implemented depending on the situation 
(author's elaboration) ......................................................................................... 74 
Table 3 - Data about the company, possible disruption, and resilience strategies 
(author's elaboration) ......................................................................................... 75 
Table 4 - Calculation of the probability of occurrence (author's elaboration) .. 77 
Table 5 - Overall impact of the disruption differentiated by degree of severity 
and  situations  considered,  after  accounting  for  the  mitigation  effect  of  the 
strategies implemented (author's elaboration) .................................................. 78 
Table 6 - Key of notations used in formulas (author's elaboration) .................. 81 
Table 7 - Data on inventory and revenue for Carel SPA for 2019 and 2020 (Orbis 
database) ............................................................................................................. 83 
Table 8 - Data on revenue and cost of raw materials for Carel SPA for 2018 (Orbis 
database) ............................................................................................................. 84 
Table 9 - Financial data for Ocado (Kahn, 2020) .............................................. 85 
Table 10 - Data about the company, possible disruption, and specific resilience 
strategies (author's elaboration) ........................................................................ 87 
Table 11 - Amount of mitigation accounting for synergies (author's elaboration)
 ............................................................................................................................. 89 
Table 12 - Overall impact of the disruption differentiated by degree of severity 
and situations considered, after accounting for the aggregate mitigation effect of 
the strategies implemented (author's elaboration) ........................................... 90 
  

100

List of Equations

Equation 1 - Formula for computing the total impact of a given situation N for a

degree of disruption D (author's elaboration) ................................................... 79

Equation 2 - Formula for computing the average effect of disruptions given the

probability of them happening (author's elaboration) ...................................... 80

Equation 3 - Formula for minimizing the average impact (author's elaboration)

............................................................................................................................. 80

Equation 4 - Formula to aggregate the amount of mitigation for different

strategies (author's elaboration) ........................................................................ 90

101

Works Cited

Albaz, A., Mansour, T., Rida, T. & Schubert, a. J., 2020. Setting up small and

medium-size enterprises for restart and recovery, s.l.: McKinsey &

Company.

Alicke, K. & Strigel, A., 2020. Supply chain risk management is back. McKinsey

& Company, 28 January.

Arora, R., 2020. Study Finds Small Business Revenues Dropped 52% And

Payrolls Declined 54% Due To COVID Pandemic. Forbes, 7 August.

Bagshaw, E. & Powell, D., 2020. Supermarkets stockpile, toilet paper

production runs 24 hours. The Sydney Morning Herald, 3 March.

Bartels, J., 2021. GEP Research Report: Total Cost of Supply Chain Disruption

in 2020 ‘Was $4tn’. Business Information Industry Association, 27

March.

Binlot, A., 2020. Retailers Report 50 Percent Revenue Impact Due To

Coronavirus In New Survey. Forbes, 31 March.

Bureau van Dijk, 2021. Orbis. [Online]

Available at: https://orbis.bvdinfo.com/

[Accessed 28 August 2021].

Burstein, F. & Holsapple, C. W., 2008. Handbook on Decision Support

Systems 1. 1st ed. s.l.:Springer.

Carli, A., 2020. Mascherine e respiratori, ecco le fabbriche che si riconvertono.

Il Sole 24 ORE, 24 March.

102

Carrick, A., 2020. FTSE 100 plunges 3.4 per cent as Italy confirms seventh

coronavirus death. City A.M., 24 February.

Chiaramonti, D. & Maniatis, K., 2020. Security of supply, strategic storage and

Covid19: Which lessons learnt for renewable and recycled carbon fuels,

and their future role in decarbonizing transport?. Applied Energy,

271(115216).

Choi, T. Y., Rogers, D. & Vakil, a. B., 2020. Coronavirus Is a Wake-Up Call for

Supply Chain Management. Harvard Business Review, 27 March.

Chowdhury, M. M. H. & Quaddus, M., 2016. Supply chain readiness, response

and recovery for resilience. Supply Chain Management, 21(6), pp. 709-

731.

Chowdhury, P., Paul, S. K., Kaisar, S. & Moktadir, M. A., 2021. COVID-19

pandemic related supply chain studies: A systematic review.

Transportation Research Part E, Volume 148.

CNN Editorial Research, 2021. Covid-19 Pandemic Timeline Fast Facts. CNN,

9 August.

Dente, S. & Hashimoto, S., 2020. COVID-19: A pandemic with positive and

negative outcomes on resource and waste flows and stocks. Resources,

Conservation & Recycling, Volume 161.

Dumbravă, V. & Iacob, V.-S., 2013. Using Probability – Impact Matrix in

Analysis and Risk Assessment Projects. Journal of Knowledge

Management, Economics and Information Technology, Volume 3, pp. 1-

103

Duncan, I., 2020. Supply chain for medicine is strained. The Washington Post

Sunday, 3 March.

Ernst and Young, 2020. Future Consumer Index: How COVID-19 is changing

consumer behaviors. [Online]

Available at: https://www.ey.com/en_gl/consumer-products-retail/how-

covid-19-could-change-consumer-behavior

[Accessed 9 April 2021].

Fahimnia, B. & Jabbarzadeh, A., 2016. Marrying supply chain sustainability

and resilience: A match made in heaven. Transportation Research Part

E: Logistics and Transportation Review, Volume 91, pp. 306-324.

Fathollahi-Fard, A. M., Hajiaghaei-Keshteli, M. & Mirjalili, S., 2018. Multi-

objective stochastic closed-loop supply chain network design with social

considerations. Applied Soft Computing, Volume 71, pp. 505-525.

Finances Online, 2021. Supply Chain Statistics: 2020/2021 Market Share

Analysis & Data. [Online]

Available at: https://financesonline.com/supply-chain-statistics/

[Accessed 2 September 2021].

Financial Express, 2020. Likelihood of future pandemics, impact and how to

reduce pandemic risks explained. Financial Express, 6 November.

Fox-Penner, P., 2020. Will the COVID-19 Pandemic Slow the Global Shift to

Renewable Energy?. The Brink, 2 April.

104

Gaudenzi, B. & Borghesi, A., 2006. Managing risks in the supply chain using

the AHP method. The International Journal of Logistics Management,

17(1), pp. 114-136.

Goel, A., 2021. How Technology Can Help With Covid-19 Supply Chain

Disruptions. Forbes, 28 January.

GS1 US, 2021. Beyond Sustainability: A Circular Approach to Building

Supply Chain Resilience, s.l.: GS1 US.

Gunessee, S. & Subramanian, N., 2020. Ambiguity and its coping mechanisms

in supply chains lessons from the Covid-19 pandemic and natural

disasters. International Journal of Operations & Production

Management, 21 July, 40(7/8), pp. 1201-1223.

Hakovirta, M. & Denuwara, N., 2020. How COVID-19 Redefines the Concept

of Sustainability. Sustainability, 12(9).

Haren, P. & Simchi-Levi, D., 2020. How Coronavirus Could Impact the Global

Supply Chain by Mid-March. Harvard Business Review, 28 February.

ISM World, 2020. Covid-19 Resource Center. [Online]

Available at: https://www.ismworld.org/supply-management-news-and-

reports/reports/covid-19-resource-center/

[Accessed 19 April 2021].

ISM, 2020. Coronavirus outbreak in China: Impact to Supply Chain, s.l.: s.n.

Ivanov, D. & Das, A., 2020. Coronavirus (COVID-19/SARS-CoV-2) and supply

chain resilience: a research note. Int. J. Integrated Supply Management,

13(1), pp. 90-102.

105

Jabbour, A. B. L. d. S. et al., 2020. Sustainability of supply chains in the wake

of the coronavirus (COVID-19/SARS-CoV-2) pandemic: lessons and

trends. Modern Supply Chain Research and Applications, 2(3), pp. 117-

122.

Jacobs, F. R. & Chase, R. B., 2018. Operations and Supply Chain

Management. XV ed. s.l.:McGraw-Hill Education.

Kahn, J., 2020. British online grocer Ocado, known for its automated

warehouses, acquires two U.S. robotics companies. Fortune, 2 November.

Kolenko, S., 2014. A Brief History of Modern Supply Chain Management and

Best Practices. [Online]

Available at: https://blog.procurify.com/2014/12/17/brief-history-

modern-supply-chain-management-supply-chain-management-best-

practices/

[Accessed 2 August 2021].

Kritzinger, W. et al., 2018. Digital Twin in manufacturing: A categorical

literature review and classification. IFAC-PapersOnLine, 51(11), pp.

1016-1022.

Lee, H. L., Padmanabhan, V. & Whang, S., 1997. Information Distortion in a

Supply Chain: The Bullwhip Effect. Management Science, 43(3), pp. 546-

558.

Lierow, M., Herzog, C. & Zollneritsch, M., 2020. Coronavirus Is Rupturing

Global Supply Chains. Brink News, 26 February.

106

Linton, T. & Vakil, B., 2020. Coronavirus Is Proving We Need More Resilient

Supply Chains. Harvard Business Review, 5 March.

Logmore, 2019. A Short History of Supply Chain Management. [Online]

Available at: https://www.logmore.com/post/history-of-supply-chain-

management

[Accessed 28 July 2021].

Lummus, R. R. & Vokurka, R. J., 1999. Defining supply chain management: a

historical perspective and practical guidelines. Industrial Management &

Data Systems, 99(1), pp. 11-17.

Lund, S. et al., 2020. Risk, resilience, and rebalancing in global value chains,

s.l.: McKinsey Global Institute.

Mahler, D. G. et al., 2021. Updated estimates of the impact of COVID-19 on

global poverty: Turning the corner on the pandemic in 2021?. World

Bank, 24 June.

Miller, J. Y., 2020. WashU Expert: Coronavirus far greater threat than SARS to

global supply chain. The Source - Washington University in St. Louis, 7

February.

Nuce, M., 2021. Entrepreneurs Disrupt Wasteful Supply Chain Practices.

Entrepreneur, 4 August.

Oliveira, U. R. d., Espindola, L. S. & Marins, F. A. S., 2018. Analysis of supply

chain risk management researches. Gest. Prod., 25(4), pp. 671-695.

Pritchard, J., 2021. Bank Stress Tests. The Balance, 14 February.

107

Ptak, C. & Smith, C., 2016. Demand Driven Material Requirements Planning

(DDMRP). s.l.:Industrial Press, Inc.

Rahman, S. M., Kim, J. & Laratte, B., 2021. Disruption in Circularity? Impact

analysis of COVID-19 on ship recycling using Weibull tonnage estimation

and scenario analysis method. Resources, Conservation & Recycling ,

Volume 164.

Reynolds, M. & Weiss, S., 2020. How coronavirus started and what happens

next, explained. Wired, 27 May.

Rice, J. B. J., 2020. Prepare Your Supply Chain for Coronavirus. Harvard

Business Review, 27 February.

Santos, D., 2020. How Retailers Globally are Responding to Coronavirus.

Aislelabs, 2 April.

Sharmaa, A., Adhikary, A. & Borah, S. B., 2020. Covid-19′s impact on supply

chain decisions: Strategic insights from NASDAQ 100 firms using Twitter

data. Journal of Business Research, Volume 117, pp. 443-449.

Simchi-Levi, D., 2015. Find the Weak Link in Your Supply Chain. Harvard

Business Review, 9 June.

Simchi-Levi, D., Schmidt, W. & Wei, Y., 2014. From Superstorms to Factory

Fires: Managing Unpredictable Supply-Chain Disruptions. Harvard

Business Review, January-February.

Simchi-Levi, D. & Simchi-Levi, E., 2020. We Need a Stress Test for Critical

Supply Chains. Harvard Business Review, 28 April.

108

Slack, N., Brandon-Jones, A. & Johnston, R., 2016. Operations Management.

VIII ed. s.l.:Pearson.

The Geography of Transport Systems, n.d. Intermodal Transport Chain.

[Online]

Available at:

https://transportgeography.org/contents/chapter5/intermodal-

transportation-containerization/intermodal-transport-chain/

[Accessed 03 August 2021].

Trautrims, A., Schleper, M. C., Cakir, M. S. & Gold, S., 2020. Survival at the

expense of the weakest? Managing modern slavery risks in supply chains

during COVID-19. Journal of Risk Research, 23(7-8), pp. 1067-1072.

Turner, B., 2020. 'Most significant crisis in the history of travel': where to now

for tourism?. The Sydney Morning Herald, 4 April.

Van der Laan, E. A., 2019. Closed Loop Supply Chain Management. In:

Operations, Logistics and Supply Chain Management. s.l.:Springer

International Publishing.

van Hoek, R., 2020. Research opportunities for a more resilient post-COVID-

19 supply chain – closing the gap between research findings and industry

practice. International Journal of Operations & Supply Chain

Management, 40(4), pp. 341-355.

Veridian, 2018. The Evolution and History of Supply Chain Management.

[Online]

Available at: https://veridian.info/history-of-supply-chain-management/

[Accessed 1 August 2021].

109

Williamson, C., 2020. The Supply Chain Of The Future Isn’t Afraid To Pivot.

Forbes, 1 September.

Wilson, M., 2020. Why COVID-19 is Different from Other Supply Chain

Disruptions. Supply & Demand Chain Executive, 2 September.

Winston, W. L. & Albright, S. C., 2019. Practical Management Science. 6th ed.

s.l.:Cengage.

Worldometer, 2021. COVID-19 CORONAVIRUS PANDEMIC. [Online]

Available at: https://www.worldometers.info/coronavirus/

[Accessed 14 August 2021].

Yeung, J., 2019. The risk of a global pandemic is growing -- and the world isn't

ready, experts say. CNN, 18 September.

Zhu, G., Chou, M. C. & Tsai, C. W., 2020. Lessons Learned from the COVID-19

Pandemic - Exposing the Shortcomings of Current Supply Chain

Operations: A Long-Term Prescriptive Offering. Sustainability, 12(14).

Zijm, H., Klumpp, M., Heragu, S. & Regattieri, A., 2019. Operations, Logistics

and Supply Chain Management: Definitions and Objectives. In:

Operations, Logistics and Supply Chain Management. s.l.:Springer

International Publishing.

0 views·113 pages

Improving the resilience of supply chains against disruptions like the Covid-19 pandemic PDF Free Download

Improving the resilience of supply chains against disruptions like the Covid-19 pandemic PDF free Download. Think more deeply and widely.

Uploaded by Tommy Kirby on 4/13/2026

/113

100%

Master’s Degree in Management -

International Management

Final Thesis

Improving the resilience of supply chains

against disruptions like the Covid-19

pandemic

Supervisor

Ch. Prof. Marco Tolotti

Graduand

Francesca Putti

Matriculation Number: 879105

Academic Year

2020/2021

This page was intentionally left blank

Table of Contents 
Introduction .......................................................................................................... 1 
The supply chain ............................................................................................ 3 
1 History ..................................................................................................... 3 
2 Definition and scope of supply chains .................................................... 6 
3 Closed-loop supply chains ..................................................................... 10 
4 Strategic Inventory Positioning  ............................................................ 13 
The Covid-19 pandemic and supply chains ................................................. 21 
1 The outbreak .......................................................................................... 21 
2 Impacts on supply chains ...................................................................... 24 
2.1 Demand Management ..................................................................... 28 
2.2 Supply management ....................................................................... 30 
2.3 Production management................................................................. 31 
2.4 Transportation and logistics management ..................................... 32 
2.5 Relationship management .............................................................. 34 
2.6 Supply chain-wide impact (causing impacts in internal, upstream, 
and downstream operations) ....................................................................... 36 
2.7 Financial management .................................................................... 39 
2.8 Sustainability management ............................................................ 40 
Building resilience  ....................................................................................... 42 
1 Risk management .................................................................................. 44 

1.1 Assessment ...................................................................................... 45 
1.2 Prevention ....................................................................................... 49 
1.3 Mitigation ........................................................................................ 50 
1.4 Recovery .......................................................................................... 51 
2 Resilience strategies .............................................................................. 52 
2.1 Assessing the supplier network structure ...................................... 52 
2.2 Stress testing ................................................................................... 56 
2.3 Diversifying sourcing ...................................................................... 58 
2.4 Increasing stock .............................................................................. 59 
2.5 Reshoring and domestic production .............................................. 60 
2.6 Technological advancements .......................................................... 62 
2.7 Increasing circularity ...................................................................... 66 
2.8 Altering production ......................................................................... 67 
Recommending resilience strategies through decision support systems ... 69 
1 Decision trees ......................................................................................... 69 
2 The general model ................................................................................. 71 
3 The model considering synergies .......................................................... 82 
4 Additional considerations ..................................................................... 91 
Conclusions ......................................................................................................... 95 
References ........................................................................................................... 97 
Works Cited....................................................................................................... 101 